Ancient Greek Theater Research Paper

Devon Whitaker Research paper Mrs. Smith December 4, 2013 The theatre of Ancient Greece, or ancient Greek drama, is a theatrical culture that flourished in ancient Greece between 550 BC and 220 BC. The city-state of Athens, which became a significant cultural, political, and military power during this period, was its center, where it was institutionalized as part of a festival called the Dionysia, which honored the god Dionysus. Tragedy, comedy, and the satire play were the three dramatic genres to emerge there. Athens exported the festival to its numerous colonies and allies in rder to promote a common cultural identity.Clothing in ancient Greece primarily consisted of the chiton, peplos, himation, and chlamys. While no clothes have survived from this period, descriptions exist in contemporary accounts and artistic depictions. Clothes were mainly homemade, and often served many purposes. Despite popular imagination and media depictions of all-white clothing, elaborate design and bri ght colors were favored. Ancient Greek clothing consisted of lengths of linen or wool fabric, which generally was rectangular. Clothes were secured with ornamental lasps or pins, and a belt, sash, or girdle might secure the waist.Men's robes went down to their knees, whereas women's went down to their ankles. The choruses were only men, as were the actors and the audience. The plays originally had a chorus of up to 50 people who performed the plays accompanied by music, beginning in the morning and lasting until the evening. They had to be citizens of Athens, which only applied to free-born men, with few special cases. The actors wore masks, so that the people would know which persona the actor played. The theatres were built on a very large scale to accommodate the large number of eople on stage, as well as the large number of people in the audience.Actors' voices needed to be heard throughout the theatre, including the very top row of seats. In 465 BC, the Theaters began using a b ackdrop, which hung behind the orchestra, which also served as an area where actors could change their costumes. It was known as the scene. In 425 BC a stone scene wall, called a paraskenia, became a common replacement to scene in the theatres Work Cited: â€Å"Theatre of Ancient Greece. † n. d. web. 04 Dec. 2013. – Simple English Wikipedia, the Free Encyclopedia. N. p. ,

Hot zone

There are a number of characters in this book, choose one and tell us why you would want to be that person. Throughout this book we are introduced to many interesting and riveting characters but in my opinion one of the best characters in this book would have to be Major Nancy Jaax. She was a veterinarian in the Army, and her work at Fort Detrick in Maryland often took her away from her children. Consequently, she often made up batches of meals in advance so they could easily be thawed and reheated in the microwave (Preston, 1994). She and her husband, Jerry, met in college and both became veterinarians.They eventually entered the military together as members of the Army's Veterinary Corps(Preston, 1994). They lived in Maryland with their two children, Jason and Jaime, and various pets. Nancy's work took her away from her family in other respects, as well, and she missed saying goodbye to her dying father because she felt that leaving during the decontamination mission would be a der eliction of duty (Preston, 1994). Nancy Jaax had to fight to get into the pathology group at the United States Army Medical Research Institute of Infectious Diseases (Preston, 1994).At that time, her status as a married female† made other people feel that she was unqualified for the Job and that she would panic in a dire situation. The military at that time was still a very male-oriented organization. In addition, Nancy had bad reactions to the vaccinations necessary to enter the program (Preston, 1994). She actually wanted to get into the Level 4, or the highest-risk part of the program, because there is no vaccine for those agents. Finally, Nancy's hands tended to move very quickly, and that made others nervous (Preston, 1994).When individuals handle sharp instruments that could be contaminated witn virus-intested b everyone wants to believe that his or ner partner is going to handle these instruments safely. Over the course of time, Nancy battled through each of these objec tions. She studied martial arts to control her movements, and at 5 feet, 4 inches, she could knock a 6-foot-tall man to the ground easily (Preston, 1994). Getting accepted into the program also included her standing up for herself to the colonel in charge of the program (Preston, 1994).She proved her desire and her competency, and by the time of the outbreak at the monkey house in Reston, Nancy Jaax had been promoted to the Chief of Pathology at USAMRIID Preston, 1994). 2. In your opinion did the government react appropriately when they decided to destroy the monkeys in Reston? Why or why not? In my opinion, yes the government did react appropriately when they decided to destroy the monkeys in Reston. The monkeys at a research facility were infected with a strain of Ebola.The military, along with the Centers for Disease Control (CDC), takes on the task of entering the monkey house and destroying the animals in an attempt to keep the virus from Jumping into the human population and c ausing a potentially worldwide crisis (Preston, 1994). The entire facility must be treated as a Hot Zone, and hundreds of monkeys are killed. Scares abound throughout the procedure: one woman's ventilated suit runs out of battery power, a monkey thought to be unconscious wakes up on the operating table while it is being euthanized and tries to bite a soldier, and tears occur in various members' space suits

Building a Portfolio Style Website

Building a Portfolio Style Website Short History and Development of HTML HTML, decrypted as a Hyper Text Markup Language. The first version of HTML appeared in 1986, in 1991 it has been significantly modified. From then until today, there have been several versions of the substantially modified. In 1995 published a second version – HTML 2.0. After the release of the second version immediately began work on the next generation of HTML. And In 1997 comes the recommendation HTML 3.2, which added a mark-up language tables, frames, images and some other important tags. The 4th version of HTML 4.01 started in 1997, recent changes appeared 24.12.1999. HTML5 It is the fifth and current version of the HTML standard; it was published in October 2014. HTML5 structure consists of a plurality of elements:    †¦    †¦ Some few examples tags and applicable attribute of HTML 1. is mandatory for the image element is mandatory for the URL of the link. This is a paragraph. Importance of CSS in web des ign and outline its evolution from CSS1 to CSS3. The first CSS specification, CSS1, became a World Wide Web Consortium (W3C) recommendation in December 1996. CSS2 came out in 1998. The work on CSS3 has been going on for years, but seems to advance very slowly So, what exactly does CSS stand for? It stands for Cascading Style Sheets — and â€Å"style sheet† refers to the document itself. Ever web browser has a default style sheet, so every web page out there is affected by at least one style sheet — the default style sheet of whatever browser the web page visitor is using — regardless whether or not the web designer applies any styles. For example, my browser’s default font style is Times New Roman, size 12, so if I visited a web page where the designer didn’t apply a style sheet of their own, I would see the web page in Times New Roman, size 12. Examples of rules created in CSS:   p{ color: #02031c; } B)   body{ background -color:   #caefc6; } Outline the four steps involved in developing a site and choice of web authoring tools available. Planning, Research Design Development Testing website Hardware and software tools you need for web design Web designing takes much more than most people think. It is about ensuring that you have all the relevant hardware and software tools required to design a good and appealing website. One of the most important tools you need a computer. A computer to be used for web design should have a large hard disk and Random Access Memory, high processing speed and large storage capacity to be able to process the large video files. The next thing you need is a server that you will use to host the website. Two servers are needed, the hardware and software servers which are important for web hosting. The relevant software for instance Dreamweaver and Microsoft front page need to be available before any progress can be made in construction of the site. How the importance of the following will affect in design a website? Target market Every website should be designed for the target audience – not just for yourself or the site owner. It is therefore very important to understand who your target audience is.

The history of National Parks in the United States Research Paper

The history of National Parks in the United States - Research Paper Example This history is also considered as the history of the people who constantly worked hard to preserve and save the land which they loved throughout their lives as United Sates’ residents. The history of United States’ national parks can be traced as far back as the discovery of Yosemite in 1851. The discovery of this place of inspiring beauty elicited events which led to the legislations that were used to protect and preserve land for future generations. White men who were members of an armed battalion entered the Yosemite Valley in 1851 in search of Indians so as to drive them away from their homeland. Those white men then named the valley â€Å"Yosemite†, believing that it was the name of the Indian tribe which lived in that valley. In 1855, James Mason led a group of white men to the discovery of the valley (Gartner 1986). After failing as a gold miner for years, James Mason believed that he could prosper by establishing and running a tourist hotel in Yosemite V alley so as to promote the scenic attractions of California. Four years later, James Mason returned to the same site, but now with a photographer. Later, other writers and artists travelled to the valley and as a result images and word concerning the valley spread fast across the US. This attracted more tourists who were specifically eager to see the beautiful valley by themselves (Gartner 1986). An editor of the New York Tribune called Horace Greeley then wrote about the valley, saying that if the county of California and other relevant authorities do not take care of the safety of the trees in the valley, then he would be sure to deplore it. He perceived that the value of the valley several years later would be very high if care and caution was taken to preserve it. Therefore, the discovery of the valley served as an important path in the history of National Parks in the United States. By late 19th century, actions of the United States to tame the land had come with devastating co nsequences. Entire species of animals had been destroyed and forests had been subjected to outrageous ravage. All these actions were committed in the name of progress. One naturalist named John Muir then expressed his concerns by categorically stating that the great wilds of the United States of America which were once boundless and inexhaustible had now become invaded and destructed completely. Within this period, there were only a handful of concerned Americans who perceived that national parks were the only structures that were considered as the important means to protect the country’s pristine places. A young politician named Theodore Roosevelt was one of the few concerned people as of that moment. Roosevelt was later to become the president of the United States of America and establish five national parks, 51 bird sanctuaries, four national game reserves, 18 monuments and 100 million acres of national forests. In 1890, there were already four national parks established d ue to the concern of the few people who were determined to preserve the environment. Despite the fact that these national parks were under the guard of the army, they were nonetheless subject to great dangers (Albright 1985). Wildlife in the park was constantly killed; park meadows were overgrazed by livestock; tourists provided means for the destruction of rocks and trees through carvings and ancient forests were not spared either. Although the congress had created the

Fast Food Rulers in China Research Paper Example | Topics and Well Written Essays - 1000 words

Fast Food Rulers in China - Research Paper Example KFC offered food items common in most Chinese restaurants ( Lroche, Kalamas &Huang, 2005). This strategic approach depicts KFC as part of the Chinese community rather than a fast food joint selling low priced westernized food. The company capitalized on small Chinese cities and the establishment of a national business with food joints spread across the country. As a result, the company cut down some of it cost due to economies of scale and distribution of risk. KFC engaged the services of Chinese hotel managers to provide advice on the food tastes. It also established partnership with local food chains and employed more Chinese to operate its emerging branches. So far, the company commands 40% stake in Chinese fast food market with 3300 food outlets in the 650 cities in China. In 1999, KFC developed a distribution chain by building warehouses and managing a fleet of distribution trucks. The trucks were fitted with refrigerators that ensured the foodstuff remain fresh while transporting them from the farm to the restaurants. Though it was an expensive affair, it was vital for the company’s rapid expansion to other cities (Schroder & McEacher, 2005) On the other hand, McDonalds a key rival of KFC holds a 16% stake in the Chinese fast food market. Its approach was far different from it competitor. It chose to stick to its core strategy adopted in the US market. MacDonald menu had no additional dishes that matched the local taste. The layouts of MacDonald’s food outlet depicted a westernized culture. Its target market was the stylish wealthy status-conscious Chinese that sought to imitate the American lifestyle. The McDonald now boasts of 2000 outlets spread across the Chinese cities. It emerged as a global leader in the fast food industry, based on sales, market capitalization, number of employees and revenues (Shen & Xiao, 2014). Its success is attributable to the quality standards the company has maintained globally

The Shawshank Redemption Movie study assignment

The Shawshank Redemption Movie study - Assignment Example Even when he is about to give up on ever getting the freedom he desires, Andy comes in and reignites that hope. He eventually gets his application granted. The importance of hope in one’s life is shown in the movie through various scenes that show the effects of hope on the lives of the prisoners. In his first night in prison, it is hope that comforts Andy. Although he is serving a life sentence, he still has hope that he will taste freedom again. It makes him calm and able to think and reflect on things thus coming up with great ideas in the process. When writing to the state senate to ask for funds to expand the prison library, hope pushes him to keep writing even when he gets no reply until finally he gets a response. The news that Tommy once met the man that killed his wife and her lover pumps new energy into Andy. His hope of getting out rises as he sees an opportunity to appeal his sentence. In addition, it is the hope of getting out someday that keeps him drilling a tunnel through the prison wall using a rock hummer for several years until he achieves his aim. Hope is also used to bring out the determination of Red to get parole despite his application being rejected repeatedly. It is hope that makes him keep applying for parole without giving up. When Brooks is released from prison, he is lost in a world he is not familiar with. He finds life hard outside prison, as he cannot adjust well. Due to a lack of hope in life, he decides to end his life by hanging himself. The verse that the warden told Andy when he first visited him in his cell was John 8:12. This verse says, â€Å"I am the Light of the world. He who follows me should not walk in darkness but will have the Light of Life.† This is the use of allusion as the warden alludes to the Bible verse to talk about his importance in the prison. He is indirectly referring to himself as the light within the prison and that whoever associates with him will get an easier time

Identify a particular issue or problem that occurs with HRD in a Essay

Identify a particular issue or problem that occurs with HRD in a country with which you are familiar - Essay Example The calculation is very simple.   The number of employees leaving in a year is calculated as a percentage of the total number of employees during the same period.   It is also known as separation rate. Employees in the beginning and closing of the year are averaged for this purpose.   However, if the monthly beginning or closing figures for the twelve months are averaged, it would be better. Some times the companies calculate a retention rate, and alternative method of calculation.   It is also called as Stability Index and worked out as below. The figures in the calculation will be unduly inflated if the replacements are frequent during the year.   For example 85% of the employees are retained.   If the vacancies caused during the year are replaced twice, the labor turnover ratio works out to 30%, if replaced only once, it works out to 15%.   Therefore retention ratio is preferred.  Ã‚   Another problem is in averaging.   If there is huge variation during the year from beginning to end or average, the ratio will be vitiated.   In a country like India, this may happen due to seasonal factors also.   For examples, if rain gods play a trick on farmers, production, consequently the employment is affected in sugar industry.  Ã‚   Michael Hanni and Mark Knold pointed out, ’construction, retail trade, administrative services, and accommodation and food services. Together, these four industries make up 35 percent of all employment, yet constitute 52.2 percent of all job separations (separations were the minimum variable of the numerator for 2006). In other words, these four industries account for a disproportionate amount of job churning.’  Labor Turnover in Utah, Source: U.S. Bureau of Census, LED data.http://jobs.utah.gov/opencms/wi/pubs/specialreports/laborturnover08.pdf We usually consider ‘Year’  as a basis for calculation.   However, year

Management Essay Example | Topics and Well Written Essays - 1000 words

Management - Essay Example Like most economies, the UK practices a system where most critical social services are provided by the government (Engineers, 2008). Education and healthcare sectors are some of the areas where the government plays very imperative roles. However, with a growing budget deficit that continues to raise concerns, austerity measures become inevitable. The United Kingdom, with its expanding public sector coupled with a growing population, has been forced to borrow severally to finance its budget deficits. Such continued borrowing may have long-term impacts on the economy considering that the debts have to be repaid. In such a case, the austerity measures advocated for by most policy advisors become necessary. It certainly becomes a plus to the many private firms across the country as new opportunities will emerge in product and service delivery. However, in light of the recent street protests against the proposals to cut government spending, several considerations emerge. Several hospitals in the UK are already bogged down by numbers courtesy of the cuts that greatly impact on the quality of social services. The education system in the country has for several years been one of the best in the world (Al-Mazrouei, 2001). This attribute emerges out of the great support that the government allocates to the basic services like education, water and healthcare. It therefore becomes imperative that several considerations be made before any major government spending is practicated. In most cases, increased taxes are normally dreaded by the public. Nevertheless, a government that effectively uses the tax money on proper policies that impact positively on the masses certainly faces little antipathy towards its social reform policies. The UK’s situation in indeed one of the most notable cases in the entire world. It demands greater government intervention which can only be attained through proper spending policies. In its bid to exercise its mandate in the establishment o f the lending base rate that guides the interest rates in England, the Bank of England’s monetary policy Committee faces myriad challenges in its approach to the whole issue (Tennant, 2009). The essence of setting the base rate is basically to ensure price stability in the country and to limit variations in the various interest rates across the country. In light of the emerging economic challenges that continue to bedevil the world, it is inevitable that economic priorities too need to change with a view of addressing the whole issue. Currently, the Committee’s main focus is keeping the interest rates at the basement levels with a view of economic expansion. In the recent years, it is indeed true that an inflation overshot has characterized the UK economy as the Banks pretence that the situation will return to normalcy in two years only help to escalate the tensions that emerge in policy circles. However, in view of this challenge, the basic challenge that confronts th e Bank of England needs to be understood. The excess price and output volatilities are issues that must be addressed in tandem with the inflationary extremities (Giuseppi, 2008). The high inflation rates in the country have greatly influenced the performance of many firms. As the borrowing rates become unbearable, most commercial firms are reconsidering their

Financial Risk Management Essay Example | Topics and Well Written Essays - 1000 words

Financial Risk Management - Essay Example This study involves a comprehensive study of the risk management policies followed by Bear Stearns and how it led to its demise. Risk Management: An Overview Risk is a term associated with any type of business entity. Without risk it would have been an easy task for managers of a company to allocate its resources in the most effective way. And with the world experiencing the global financial crisis in the year 2008, effective and efficient risk management is the key to success for any financial enterprise. The idea of risk management may differ from person to person. In case of regulators risk management is a means of control, for traders it is a means of hedging their risks and for risk managers it is a means of obtaining the highest return possible by allocating capital in the best possible way. Risk management takes into consideration the magnitude as well as the nature of risks involved. It is all about optimizing the risk-return profile of a company. Sources of risk are many and it is due to the uncertainties of future events. In today’s world banks are engaged in wide range of activities like trading of derivatives to its customers which results in exposure, finally leading to risks. Thus risk management plays a vital role in case of banks. Study and analysis of risk begins with study of Markowitz model of portfolio analysis where he defined selection of portfolios based on mean of variances in return of portfolios. Sharpe and Lintner further added to this analysis by assuming the existence of risk free assets. The rate of return of a risky asset is governed by its systematic risk and ‘beta’ is the measure of this type of risks. Next Black-Scholes model for pricing of options gives a measure of the risk of an underlying security by measuring the volatility in the form of standard deviation. Again works of Modigliani and Miller showed that value of the firm is not dependent on the capital structure of the firm. Increase of debt, leading to greater leverage in the capital structure of a firm increases the financial risk for the shareholders of the firm. This means, reengineering of capital structure of the firm would not help the firm, and the management should consider implementing strategies to increase the value of the firm economically. However, it is very difficult and possesses a challenging task for a company to implement these risk management theories practically. For any financial institutions like global banks, before taking up risk management system they must ensure that, they are up to date with their databases regarding various financial transactions within the company and are also aware about financial rates available in the outside market. Also they must have relevant statistical tools to analyze those data. Risk management policies followed by Bear Stearns Bear Stearns was once considered as one of the most efficient managers of risk but at the end it was their faulty risk management policies that l ed to their downfall. Bear Stearns most profitable division was the securitization of mortgage, which brought in almost half of the company’s revenues. Regarding mortgage securitization, the company followed a model that was vertically integrated and made profit at every step starting from originating loan, securitizing them and then selling them. These

Approaches to the Analysis of Survey Data Essay Example for Free

Approaches to the Analysis of Survey Data Essay 1. Preparing for the Analysis 1.1 Introduction This guide is concerned with some fundamental ideas of analysis of data from surveys. The discussion is at a statistically simple level; other more sophisticated statistical approaches are outlined in our guide Modern Methods of Analysis. Our aim here is to clarify the ideas that successful data analysts usually need to consider to complete a survey analysis task purposefully. An ill-thought-out analysis process can produce incompatible outputs and many results that never get discussed or used. It can overlook key findings and fail to pull out the subsets of the sample where clear findings are evident. Our brief discussion is intended to assist the research team in working systematically; it is no substitute for clear-sighted and thorough work by researchers. We do not aim to show a totally naà ¯ve analyst exactly how to tackle a particular set of survey data. However, we believe that where readers can undertake basic survey analysis, our recommendations will help and encourage them to do so better. Chapter 1 outlines a series of themes, after an introductory example. Different data types are distinguished in section 1.2. Section 1.3 looks at data structures; simple if there is one type of sampling unit involved, and hierarchical with e.g. communities, households and individuals. In section 1.4 we separate out three stages of survey data handling – exploration, analysis and archiving – which help to define expectations and procedures for different parts of the overall process. We contrast the research objectives of description or estimation (section 1.5), and of comparison  (section 1.6) and what these imply for analysis. Section 1.7 considers when results should be weighted to represent the population – depending on the extent to which a numerical value is or is not central to the interpretation of survey results. In section 1.8 we outline the coding of non-numerical responses. The use of ranked data is discussed in brief in section 1.9. In Chapter 2 we look at the ways in which researchers usually analyse survey data. We focus primarily on tabular methods, for reasons explained in section 2.1. Simple one-way tables are often useful as explained in section 2.2. Cross-tabulations (section 2.3) can take many forms and we need to think which are appropriate. Section 2.4 discusses issues about ‘accuracy’ in relation to two- and multi-way tables. In section 2.5 we briefly discuss what to do when several responses can be selected in response to one question.  © SSC 2001 – Approaches to the Analysis of Survey Data 5 Cross-tabulations can look at many respondents, but only at a small number of questions, and we discuss profiling in section 2.6, cluster analysis in section 2.7, and indicators in sections 2.8 and 2.9. 1.2 Data Types Introductory Example: On a nominal scale the categories recorded, usually counted, are described verbally. The ‘scale’ has no numerical characteristics. If a single oneway table resulting from simple summarisation of nominal (also called categorical) scale data contains frequencies:Christian Hindu Muslim Sikh Other 29 243 117 86 25 there is little that can be done to present exactly the same information in other forms. We could report highest frequency first as opposed to alphabetic order, or reduce the information in some way e.g. if one distinction is of key importance compared to the others:Hindu Non-Hindu 243 257 On the other hand, where there are ordered categories, the sequence makes sense only in one, or in exactly the opposite, order:Excellent Good Moderate Poor Very Bad 29 243 117 86 25 We could reduce the information by combining categories as above, but also we can summarise, somewhat numerically, in various ways. For example, accepting a degree of arbitrariness, we might give scores to the categories:Excellent Good Moderate Poor Very Bad 5 4 3 2 1 and then produce an ‘average score’ – a numerical indicator – for the sample of:29 Ãâ€" 5 + 243 Ãâ€" 4 + 117 Ãâ€" 3 + 86 Ãâ€" 2 + 25 Ãâ€" 1 29 + 243 + 117 + 86 + 25 = 3.33 This is an analogue of the arithmetical calculation we would do if the categories really were numbers e.g. family sizes. 6  © SSC 2001 – Approaches to the Analysis of Survey Data The same average score of 3.33 could arise from differently patterned data e.g. from rather more extreme results:Excellent Good Moderate Poor Very Bad 79 193 117 36 75 Hence, as with any other indicator, this ‘average’ only represents one feature of the data and several summaries will sometimes be needed. A major distinction in statistical methods is between quantitative data and the other categories exemplified above. With quantitative data, the difference between the values from two respondents has a clearly defined and incontrovertible meaning e.g. â€Å"It is 5C ° hotter now than it was at dawn† or â€Å"You have two more children than your sister†. Commonplace statistical methods provide many well-known approaches to such data, and are taught in most courses, so we give them only passing attention here. In this guide we focus primarily on the other types of data, coded in number form but with less clear-cut numerical meaning, as follows. Binary – e.g. yes/no data – can be coded in 1/0 form; while purely categorical or nominal data – e.g. caste or ethnicity – may be coded 1, 2, 3†¦ using numbers that are just arbitrary labels and cannot be added or subtracted. It is also common to have ordered categorical data, where items may be rated Excellent, Good, Poor, Useless, or responses to attitude statements may be Strongly agree, Agree, Neither agree nor disagree, Disagree, Strongly disagree. With ordered categorical data the number labels should form a rational sequence, because they have some numerical meaning e.g. scores of 4, 3, 2, 1 for Excellent through to Useless. Such data supports limited quantitative analysis, and is often referred to by statisticians as ‘qualitative’ – this usage does not imply that the elicitation procedure must satisfy a purist’s restrictive perception of what constitutes qualitative research methodology. 1.3 Data Structure SIMPLE SURVEY DATA STRUCTURE: the data from a single-round survey, analysed with limited reference to other information, can often be thought of as a ‘flat’ rectangular file of numbers, whether the numbers are counts/measurements, or codes, or a mixture. In a structured survey with numbered questions, the flat file has a column for each question, and a row for each respondent, a convention common to almost all standard statistical packages. If the data form a perfect rectangular grid with a number in every cell, analysis is made relatively easy, but there are many reasons why this will not always be the case and flat file data will be incomplete or irregular. Most importantly:-  © SSC 2001 – Approaches to the Analysis of Survey Data 7 †¢ Surveys often involve ‘skip’ questions where sections are missed out if irrelevant e.g. details of spouse’s employment do not exist for the unmarried. These arise legitimately, but imply different subsets of people respond to different questions. ‘Contingent questions’, where not everyone ‘qualifies’ to answer, often lead to inconsistent-seeming results for this reason. If the overall sample size is just adequate, the subset who ‘qualify’ for a particular set of contingent questions may be too small to analyse in the detail required. †¢ If some respondents fail to respond to some questions (item non-response) there will be holes in the rectangle. Non-informative non-response occurs if the data is missing for a reason unrelated to the true answers e.g. the interviewer turned over two pages instead of one! Informative non-response means that the absence of an answer itself tells you something, e.g. you are almost sure that the missing income value will be one of the highest in the community. A little potentially informative non-response may be ignorable, if there is plenty of data. If data are sparse or if informative  non-response is frequent, the analysis should take account of what can be inferred from knowing that there are informative missing values. HIERARCHICAL DATA STRUCTURE: another complexity of survey data structure arises if the data are hierarchical. A common type of hierarchy is where a series of questions is repeated say for each child in the household, and combined with a household questionnaire, and maybe data collected at community level. For analysis, we can create a rectangular flat file, at the ‘child level’, by repeating relevant household information in separate rows for each child. Similarly, we can summarise information for the children in a household, to create a ‘household level’ analysis file. The number of children in the household is usually a desirable part of the summary; this â€Å"post-stratification† variable can be used to produce sub-group analyses at household level separating out households with different numbers of child members. The way the sampling was done can have an effect on interpretation or analysis of a hierarchical study. For example if children were chosen at random, households with more children would have a greater chance of inclusion and a simple average of the household sizes would be biased upwards: it should be corrected for selection probabilities. Hierarchical structure becomes important, and harder to handle, if there are many levels where data are collected e.g. government guidance and allocations of resource, District Development Committee interpretations of the guidance, Village Task Force selections of safety net beneficiaries, then households and individuals whose vulnerabilities and opportunities are affected by targeting decisions taken at higher levels in the hierarchy. In such cases, a relational database reflecting the hierarchical 8  © SSC 2001 – Approaches to the Analysis of Survey Data structure is a much more desirable way than a spreadsheet to define and retain the inter-relationships between levels, and to create many analysis files at different levels. Such issues are described in the guide The Role of a Database Package for Research Projects. Any one of the analysis files   may be used as we discuss below, but any such study will be looking at one facet of the structure, and several analyses will have to be brought together for an overall interpretation. A more sophisticated approach using multi-level modelling, described in our guide on Modern Methods of Analysis, provides a way to look at several levels together. 1.4 Stages of Analysis It is often worth distinguishing the three stages of exploratory analysis, deriving the main findings, and archiving. EXPLORATORY DATA ANALYSIS (EDA) means looking at the data files, maybe even before all the data has been collected and entered, to get an idea of what is there. It can lead to additional data collection if this is seen to be needed, or savings by stopping collecting data when a conclusion is already clear, or existing results prove worthless. It is not assumed that results from EDA are ready for release as study findings. †¢ EDA usually overlaps with data cleaning; it is the stage where anomalies become evident e.g. individually plausible values may lead to a way-out point when combined with other variables on a scatterplot. In an ideal situation, EDA would end with confidence that one has a clean dataset, so that a single version of the main datafiles can be finalised and ‘locked’ and all published analyses derived from a single consistent form of ‘the data’. In practice later stages of analysis often produce additional queries about data values. †¢ Such exploratory analysis will also show up limitations in contingent questions e.g. we might find we don’t have enough currently married women to analyse their income sources separately by district. EDA should include the final reconciliation of analysis ambitions with data limitations. †¢ This phase can allow the form of analysis to be tried out and agreed, developing analysis plans and program code in parallel with the final data collection, data entry and checking. Purposeful EDA allows the subsequent stage of deriving the main findings to be relatively quick, uncontroversial, and well organised. DERIVING THE MAIN FINDINGS: the second stage will  ideally begin with a clear-cut clean version of the data, so that analysis files are consistent with one another, and any inconsistencies, e.g. in numbers included, can be clearly explained. This is the stage we amplify upon, later in this guide. It should generate the summary  © SSC 2001 – Approaches to the Analysis of Survey Data 9 findings, relationships, models, interpretations and narratives, and recommendations that research users will need to begin utilising the results. first Of course one needs to allow time for ‘extra’ but usually inevitable tasks such as:†¢ follow-up work to produce further more detailed findings, e.g. elucidating unexpected results from the pre-planned work. †¢ a change made to the data, each time a previously unsuspected recording or data entry error comes to light. Then it is important to correct the database and all analysis files already created that involve the value to be corrected. This will mean repeating analyses that have already been done using, but not revealing, the erroneous value. If that analysis was done â€Å"by mouse clicking† and with no record of the steps, this can be very tedious. This stage of work is best undertaken using software that can keep a log: it records the analyses in the form of program instructions that can readily and accurately be re-run. ARCHIVING means that data collectors keep, perhaps on CD, all the non-ephemeral material relating to their efforts to acquire information. Obvious components of such a record include:(i) data collection instruments, (ii) raw data, (iii) metadata recording the what, where, when, and other identifiers of all variables, (iv) variable names and their interpretations, and labels corresponding to values of categorical variables, (v) query programs used to extract analysis files from the database, (vi) log files  defining the analyses, and (vii) reports. Often georeferencing information, digital photographs of sites and scans of documentary material are also useful. Participatory village maps, for example, can be kept for reference as digital photographs. Surveys are often complicated endeavours where analysis covers only a fraction of what could be done. Reasons for developing a good management system, of which the archive is part, include:†¢ keeping the research process organised as it progresses; †¢ satisfying the sponsor’s (e.g. DFID’s) contractual requirement that data should be available if required by the funder or by legitimate successor researchers; †¢ permitting a detailed re-analysis to authenticate the findings if they are questioned; †¢ allowing a different breakdown of results e.g. when administrative boundaries are redefined; †¢ linking several studies together, for instance in longer-term analyses carrying baseline data through to impact assessment. 10  © SSC 2001 – Approaches to the Analysis of Survey Data 1.5 Population Description as the Major Objective In the next section we look at the objective of comparing results from sub-groups, but a more basic aim is to estimate a characteristic like the absolute number in a category of proposed beneficiaries, or a relative number such as the prevalence of HIV seropositives. The estimate may be needed to describe a whole population or sections of it. In the basic analyses discussed below, we need to bear in mind both the planned and the achieved sampling structure. Example: Suppose ‘before’ and ‘after’ surveys were each planned to have a 50:50 split of urban and rural respondents. Even if we achieved 50:50 splits, these would need some manipulation if we wanted to generalise the results to represent an actual population split of 70:30 urban:rural. Say we wanted to assess the change from ‘before’ to ‘after’ and the achieved samples were in fact split 55:45 and 45:55. We would have to correct the  results carefully to get a meaningful estimate of change. Samples are often stratified i.e. structured to capture and represent particular segments of the target population. This may be much more sophisticated than the urban/rural split in the previous paragraph. Within-stratum summaries serve to describe and characterise each of these parts individually. If required by the objectives, overall summaries, which put together the strata, need to describe and characterise the whole population. It may be fine to treat the sample as a whole and produce simple, unweighted summaries if (i) we have set out to sample the strata proportionately, (ii) we have achieved this, and (iii) there are no problems due to hierarchical structure. Nonproportionality arises from various quite distinct sources, in particular:†¢ Case A: often sampling is disproportionate across strata by design, e.g. the urban situation is more novel, complex, interesting or accessible, and gets greater coverage than the fraction of the population classed as rural. †¢ Case B : sometimes particular strata are bedevilled with high levels of nonresponse, so that the data are not proportionate to stratum sizes, even when the original plan was that they should be. If we ignore non-proportionality, a simple-minded summary over all cases is not a proper representation of the population in these instances.  The ‘mechanistic’ response to ‘correct’ both the above cases is (1) to produce withinstratum results (tables or whatever), (2) to scale the numbers in them to represent the true population fraction that each stratum comprises, and then (3) to combine the results.  © SSC 2001 – Approaches to the Analysis of Survey Data 11 There is often a problem with doing this in case B, where non-response is an important part of the disproportionality: the reasons why data are missing from particular strata often correspond to real differences in the behaviour of respondents, especially those omitted or under-sampled, e.g. â€Å"We had very good response rates everywhere except in the north. There a high proportion of the population are nomadic, and we largely failed to find them.† Just  scaling up data from settled northerners does not take account of the different lifestyle and livelihood of the missing nomads. If you have largely missed a complete category, it is honest to report partial results making it clear which categories are not covered and why. One common ‘sampling’ problem arises when a substantial part of the target population is unwilling or unable to cooperate, so that the results in effect only represent a limited subset – those who volunteer or agree to take part. Of course the results are biased towards e.g. those who command sufficient resources to afford the time, or e.g. those who habitually take it upon themselves to represent others. We would be suspicious of any study which appeared to have relied on volunteers, but did not look carefully at the limits this imposed on the generalisability of the conclusions. If you have a low response rate from one stratum, but are still prepared to argue that the data are somewhat representative, the situation is at the very least uncomfortable. Where you have disproportionately few responses, the multipliers used in scaling up to ‘represent’ the stratum will be very high, so your limited data will be heavily weighted in the final overall summary. If there is any possible argument that these results are untypical, it is worthwhile to think carefully before giving them extra prominence in this way. 1.6 Comparison as the Major Objective One sound reason for disproportionate sampling is that the main objective is a comparison of subgroups in the population. Even if one of two groups to be compared is very small, say 10% of the total number in the population, we now want roughly equally many observations from each subgroup, to describe both groups roughly equally accurately. There is no point in comparing a very accurate set of results from one group with a very vague, ill-defined description of the other; the comparison is at least as vague as the worse description. The same broad principle applies whether the comparison is a wholly quantitative one looking at the difference in means of a numerical measure between groups, or a much looser verbal comparison e.g. an assessment of differences in pattern across a range of cross-tabulations. 12  © SSC 2001 – Approaches to the Analysis of Survey Data If for a subsidiary objective we produce an overall summary giving ‘the general picture’ of which both groups are part, 50:50 sampling may need to be re-weighted 90:10 to produce a quantitative overall picture of the sampled population. The great difference between true experimental approaches and surveys is that experiments usually involve a relatively specific comparison as the major objective, while surveys much more often do not. Many surveys have multiple objectives, frequently ill defined, often contradictory, and usually not formally prioritised. Along with the likelihood of some non-response, this tends to mean there is no sampling scheme which is best for all parts of the analysis, so various different weighting schemes may be needed in the analysis of a single survey. 1.7 When Weighting Matters Several times in the above we have discussed issues about how survey results may need to be scaled or weighted to allow for, or ‘correct for’, inequalities in how the sample represents the population. Sometimes this is of great importance, sometimes not. A fair evaluation of survey work ought to consider whether an appropriate tradeoff has been achieved between the need for accuracy and the benefits of simplicity. If the objective is formal estimation, e.g. of total population size from a census of a sample of communities, we are concerned to produce a strictly numerical answer, which we would like to be as accurate as circumstances allow. We should then correct as best we can for a distorted representation of the population in the sample. If groups being formally compared run across several population strata, we should try to ensure the comparison is fair by similar corrections, so that the groups are compared on the basis of consistent samples. In these cases we have to face up to problems such as unusually large weights attached to poorly-responding strata, and we may need to investigate the extent to which the final answer is dubious because of sensitivity to results from such subsamples. Survey findings are often used in ‘less numerical’ ways, where it may not be so important to achieve accurate weighting e.g. â€Å"whatever varieties they grow for sale, a large majority of farm households in Sri Lanka prefer traditional red rice varieties for home consumption because they prefer their flavour†. If this is a clear-cut finding which accords with other information, if it is to be used for a simple decision process, or if it is an interim finding which will prompt further investigation, there is a lot to be said for keeping the analysis simple. Of course it saves time and money. It makes the process of interpretation of the findings more accessible to those not very involved in the study. Also, weighting schemes depend on good information to create the weighting factors and this may be hard to pin down.  © SSC 2001 – Approaches to the Analysis of Survey Data 13 Where we have worryingly large weights, attaching to small amounts of doubtful information, it is natural to want to put limits on, or ‘cap’, the high weights, even at the expense of introducing some bias, i.e. to prevent any part of the data having too much impact on the result. The ultimate form of capping is to express doubts about all the data, and to give equal weight to every observation. The rationale, not usually clearly stated, even if analysts are aware they have done this, is to minimise the maximum weight given to any data item. This lends some support to the common practice of analysing survey data as if they were a simple random sample from an unstructured population. For ‘less numerical’ usages, this may not be particularly problematic as far as simple description is concerned. Of course it is wrong – and may be very misleading – to follow this up by calculating standard deviations and making claims of accuracy about the results which their derivation will not sustain! 1.8 Coding We recognise that purely qualitative researchers may prefer to use qualitative analysis methods and software, but where open-form and other verbal responses occur alongside numerical data it is often sensible to use a quantitative tool. From the statistical viewpoint, basic coding implies that we have material, which can be put into nominal-level categories. Usually this is recorded in verbal or pictorial form, maybe on audio- or videotape, or written down by interviewers or self-reported. We would advocate computerising the raw data, so it is archived. The following refers to extracting codes, usually describing the routine comments, rather than unique individual ones which can be used for subsequent qualitative analysis. By scanning the set of responses, themes are developed which reflect the items noted in the material. These should reflect the objectives of the activity. It is not necessary to code rare, irrelevant or uninteresting material. In the code development phase, a large enough range of the responses is scanned to be reasonably sure that commonly occurring themes have been noted. If previous literature, or theory, suggests other themes, these are noted too. Ideally, each theme is broken down into unambiguous, mutually exclusive and exhaustive, categories so that any response segment can be assigned to just one, and assigned the corresponding code value. A ‘codebook’ is then prepared where the categories are listed and codes assigned to them. Codes do not have to be consecutive numbers. It is common to think of codes as presence/absence markers, but there is no intrinsic reason why they should not be graded as ordered categorical variables if appropriate, e.g. on a scale such as fervent, positive, uninterested/no opinion, negative. 14  © SSC 2001 – Approaches to the Analysis of Survey Data The entire body of material is then reviewed and codes are recorded. This may be in relevant places on questionnaires or transcripts. Especially when looking at ‘new’ material not used in code development, extra items may arise and need to be added to the codebook. This may mean another pass through material already reviewed, to add new codes e.g. because a  particular response is turning up more than expected. From the point of view of analysis, no particular significance attaches to particular numbers used as codes, but it is worth bearing in mind that statistical packages are usually excellent at sorting, selecting or flagging, for example, ‘numbers between 10 and 19’ and other arithmetically defined sets. If these all referred to a theme such as ‘forest exploitation activities of male farmers’ they could easily be bundled together. It is of course impossible to separate out items given the same code, so deciding the right level of coding detail is essential at an early stage in the process. When codes are analysed, they can be treated like other nominal or ordered categorical data. The frequencies of different types of response can be counted or cross-tabulated. Since they often derive from text passages and the like, they are often particularly well-adapted for use in sorting listings of verbal comments – into relevant bundles for detailed non-quantitative analysis. 1.9 Ranking Scoring A common means of eliciting data is to ask individuals or groups to rank a set of options. The researchers’ decision to use ranks in the first place means that results are less informative than scoring, especially if respondents are forced to choose between some nearly-equal alternatives and some very different ones. A British 8-year-old offered baked beans on toast, or fish and chips, or chicken burger, or sushi with hot radish might rank these 1, 2, 3, 4 but score them 9, 8.5, 8, and 0.5 on a zero to ten scale! Ranking is an easy task where the set of ranks is not required to contain more than about four or five choices. It is common to ask respondents to rank, say, their best four from a list of ten, with 1 = best, etc. Accepting a degree of arbitrariness, we would usually replace ranks 1, 2, 3, 4, and a string of blanks by pseudo-scores 4, 3, 2, 1, and a string of zeros, which gives a complete array of numbers we can summarise – rather than a sparse array where we don’t know how to handle the blanks. A project output paper†  available on the SSC website explores this in more detail. †  Converting Ranks to Scores for an ad hoc Assessment of Methods of Communication Available to Farmers by Savitri Abeyasekera, Julie  Lawson-Macdowell Ian Wilson. This is an output from DFID-funded work under the Farming Systems Integrated Pest Management Project, Malawi and DFID NRSP project R7033, Methodological Framework for Combining Qualitative and Quantitative Survey Methods.  © SSC 2001 – Approaches to the Analysis of Survey Data 15 Where the instructions were to rank as many as you wish from a fixed, long list, we would tend to replace the variable length lists of ranks with scores. One might develop these as if respondents each had a fixed amount, e.g. 100 beans, to allocate as they saw fit. If four were chosen these might be scored 40, 30, 20, 10, or with five chosen 30, 25, 20, 15, 10, with zeros again for unranked items. These scores are arbitrary e.g. 40, 30, 20, 10 could instead be any number of choices e.g. 34, 28, 22, 16 or 40, 25, 20, 15; this reflects the rather uninformative nature of rankings, and the difficulty of post hoc construction of information that was not elicited effectively in the first place. Having reflected and having replaced ranks by scores we would usually treat these like any other numerical data, with one change of emphasis. Where results might be sensitive to the actual values attributed to ranks, we would stress sensitivity analysis more than with other types of numerical data, e.g. re-running analyses with (4, 3, 2, 1, 0, 0, †¦) pseudo-scores replaced by (6, 4, 2, 1, 0, 0 , †¦). If the interpretations of results are insensitive to such changes, the choice of scores is not critical. 16  © SSC 2001 – Approaches to the Analysis of Survey Data 2. Doing the Analysis 2.1 Approaches Data listings are readily produced by database and many statistical packages. They are generally on a case-by-case basis, so are particularly suitable in  EDA as a means of tracking down odd values, or patterns, to be explored. For example, if material is in verbal form, such a listing can give exactly what every respondent was recorded as saying. Sorting these records – according to who collected them, say – may show up great differences in field workers’ aptitude, awareness or approach. Data listings can be an adjunct to tabulation: in Excel, for example, the Drill Down feature allows one to look at the data from individuals who appear together in a single cell. There is a place for the use of graphical methods, especially for presentational purposes, where simple messages need to be given in easily understood, and attentiongrabbing form. Packages offer many ways of making results bright and colourful, without necessarily conveying more information or a more accurate understanding. A few basic points are covered in the guide on Informative Presentation of Tables, Graphs and Statistics. Where the data are at all voluminous, it is a good idea selectively to tabulate most ‘qualitative’ but numerically coded data i.e. the binary, nominal or ordered categorical types mentioned above. Tables can be very effective in presentations if stripped down to focus on key findings, crisply presented. In longer reports, a carefully crafted, well documented, set of cross-tabulations is usually an essential component of summary and comparative analysis, because of the limitations of approaches which avoid tabulation:†¢ Large numbers of charts and pictures can become expensive, but also repetitive, confusing and difficult to use as a source of detailed information. †¢ With substantial data, a purely narrative full description will be so long-winded and repetitive that readers will have great difficulty getting a clear picture of what the results have to say. With a briefer verbal description, it is difficult not to be overly selective. Then the reader has to question why a great deal went into collecting data that merits little description, and should question the impartiality of the reporting. †¢ At the other extreme, some analysts will skip or skimp the tabulation stage and move rapidly to complex statistical modelling. Their findings are just as much to be distrusted! The models may be based on preconceptions rather than evidence, they may fit badly and conceal important variations in the underlying patterns.  © SSC 2001 – Approaches to the Analysis of Survey Data 17 †¢ In terms of producing final outputs, data listings seldom get more than a place in an appendix. They are usually too extensive to be assimilated by the busy reader, and are unsuitable for presentation purposes. 2.2 One-Way Tables The most straightforward form of analysis, and one that often supplies much of the basic information need, is to tabulate results, question by question, as ‘one-way tables’. Sometimes this can be done using an original questionnaire and writing on it the frequency or number of people who ‘ticked each box’. Of course this does not identify which respondents produced particular combinations of responses, but this is often a first step where a quick and/or simple summary is required. 2.3 Cross-Tabulation: Two-Way Higher-Way Tables At the most basic level, cross-tabulations break down the sample into two-way tables showing the response categories of one question as row headings, those of another question as column headings. If for example each question has five possible answers the table breaks the total sample down into 25 subgroups. If the answers are subdivided e.g. by sex of respondent, there will be one three-way table, 5x5x2, probably shown on the page as separate two-way tables for males and for females. The total sample size is now split over 50 categories and the degree to which the data can sensibly be disaggregated will be constrained by the total number of respondents represented. There are usually many possible two-way tables, and even more three-way tables. The main analysis needs to involve careful thought as to which ones are necessary, and how much detail is needed. Even after deciding that we want some cross-tabulation with categories of ‘question J’ as rows and ‘question K’ as columns, there are several other  decisions to be made: †¢ The number in the cells of the table may be just the frequency i.e. the number of respondents who gave that combination of answers. This may be rephrased as a proportion or a percentage of the total. Alternatively, percentages can be scaled so they total 100% across each row or down each column, so as to make particular comparisons clearer. †¢ The contents of a cell can equally well be a statistic derived from one or more other questions e.g. the proportion of the respondents falling in that cell who were economically-active women. Often such a table has an associated frequency table to show how many responses went in to each cell. If the cell frequencies represent 18  © SSC 2001 – Approaches to the Analysis of Survey Data small subsamples the results can vary wildly, just by chance, and should not be over-interpreted. †¢ Where interest focuses mainly on one ‘area’ of a two-way table it may be possible to combine rows and columns that we don’t need to separate out, e.g. ruling party supporters vs. supporters of all other parties. This simplifies interpretation and presentation, as well as reducing the impact of chance variations where there are very small cell counts. †¢ Frequently we don’t just want the cross-tabulation for ‘all respondents’. We may want to have the same table separately for each region of the country – described as segmentation – or for a particular group on whom we wish to focus such as ‘AIDS orphans’ – described as selection. †¢ Because of varying levels of success in covering a population, the response set may end up being very uneven in its coverage of the target population. Then simply combining over the respondents can mis-represent the intended population. It may be necessary to show the patterns in tables, sub-group by sub-group to convey the whole picture. An alternative, discussed in Part 1, is to weight up the results from the sub-groups to give a fair representation of the whole. 2.4 Tabulation the Assessment of Accuracy Tabulation is usually purely descriptive, with limited effort made to assess the ‘accuracy’ of the numbers tabulated. We caution that confidence intervals are sometimes very wide when survey samples have been disaggregated into various subgroups: if crucial decisions hang on a few numbers it may well be worth putting extra effort into assessing – and discussing – how reliable these are. If the uses intended for various tables are not very numerical or not very crucial, it is likely to cause unjustifiable delay and frustration to attempt to put formal measures of precision on the results. Usually, the most important considerations in assessing the ‘quality’ or ‘value’ or ‘accuracy’ of results are not those relating to ‘statistical sampling variation’, but those which appraise the following factors and their effects:†¢ evenness of coverage of the target (intended) population †¢ suitability of the sampling scheme reviewed in the light of field experience and findings †¢ sophistication and uniformity of response elicitation and accuracy of field recording †¢ efficacy of measures to prevent, compensate for, and understand non-response †¢ quality of data entry, cleaning and metadata recording †¢ selection of appropriate subgroups in analysis  © SSC 2001 – Approaches to the Analysis of Survey Data 19 If any of the above factors raises important concerns, it is necessary to think hard about the interpretation of ‘statistical’ measures of precision such as standard errors. A factor that has uneven effects will introduce biases, whose size and detectability ought to be dispassionately appraised and reported with the conclusions. Inferential statistical procedures can be used to guide generalisations from the sample to the population, where a  survey is not badly affected by any of the above. Inference addresses issues such as whether apparent patterns in the results have come about by chance or can reasonably be taken to reflect real features of the population. Basic ideas are reviewed in Understanding Significance: the Basic Ideas of Inferential Statistics. More advanced approaches are described in Modern Methods of Analysis. Inference is particularly valuable, for instance, in determining the appropriate form of presentation of survey results. Consider an adoption study, which examined socioeconomic factors affecting adoption of a new technology. Households are classified as male or female headed, and the level of education and access to credit of the head is recorded. At its most complicated the total number of households in the sample would be classified by adoption, gender of household head, level of education and access to credit resulting in a 4-way table. Now suppose, from chi-square tests we find no evidence of any relationship between adoption and education or access to credit. In this case the results of the simple twoway table of adoption by gender of household head would probably be appropriate. If on the other hand, access to credit were the main criterion affecting the chance of adoption and if this association varied according to the gender of the household head, the simple two-way table of adoption by gender would no longer be appropriate and a three-way table would be necessary. Inferential procedures thus help in deciding whether presentation of results should be in terms of one-way, two-way or higher dimensional tables. Chi-square tests are limited to examining association in two-way tables, so have to be used in a piecemeal fashion for more complicated situations like that above. A more general way to examine tabulated data is to use log-linear models described in Modern Methods of Analysis. 2.5 Multiple Response Data Surveys often contain questions where respondents can choose a number of relevant responses, e.g. 20  © SSC 2001 – Approaches to the Analysis of Survey Data If you are not using an improved fallow on any of your land, please tick from the list below, any reasons that apply to you:(i) Don’t have any land of my own (ii) Do not have any suitable crop for an improved fallow (iii) Can not afford to buy the seed or plants (iv) Do not have the time/labour There are three ways of computerising these data. The simplest is to provide as many columns as there are alternatives. This is called a multiple dichotomy†, because there is a yes/no (or 1/0) response in each case indicating that the respondent ticked/did not tick each item in the list. The second way is to find the maximum number of ticks from anyone and then have this number of columns, entering the codes for ticked responses, one per column. This is known as â€Å"multiple response† data. This is a useful method if the question asks respondents to put the alternatives in order of importance, because the first column can give the most important reason, and so on. A third method is to have a separate table for the data, with just 2 columns. The first identifies the person and the second gives their responses. There are as many rows of data as there are reasons. There is no entry for a  person who gives no reasons. Thus, in this third method the length of the columns is equal to the number of responses rather than the number of respondents. If there are follow-up questions about each reason, the third method above is the obvious way to organise the data, and readers may identify the general concept as being that of data at another level, i.e. the reason level. More information on organising this type of data is provided in the guide The Role of a Database Package for Research Projects. Essentially such data are analysed by building up counts of the numbers of mentions of each response. Apart from SPSS, few standard statistics packages have any special facilities for processing multiple response and multiple dichotomy data. Almost any package can be used with a little ingenuity, but working from first principles is a timeconsuming business. On our web site we describe how Excel may be used. 2.6 Profiles Usually the questions as put to respondents in a survey need to represent ‘atomic’ facets of an issue, expressed in concrete terms and simplified as much as possible, so that there is no ambiguity and so they will be consistently interpreted by respondents.  © SSC 2001 – Approaches to the Analysis of Survey Data 21 Basic cross-tabulations are based on reporting responses to such individual questions and are therefore narrowly issue-specific. A rather different approach is needed if the researchers’ ambitions include taking an overall view of individual, or small groups’, responses as to their livelihood, say. Cross-tabulations of individual questions are not a sensible approach to ‘people-centred’ or ‘holistic’ summary of results. Usually, even when tackling issues a great deal less complicated than livelihoods, the more important research outputs are ‘complex molecules’ which bring together  responses from numerous questions to produce higher-level conclusions described in more abstract terms. For example several questions may each enquire whether the respondent follows a particular recommendation, whereas the output may be concerned with overall ‘compliance’ – the abstract concept behind the questioning. A profile is a description synthesising responses to a range of questions, perhaps in terms of a set of abstract nouns like compliance. It may describe an individual, cluster of respondents or an entire population. One approach to discussing a larger concept is to produce numerous cross-tabulations reflecting actual questions and to synthesise their information content verbally. This tends to lose sight of the ‘profiling’ element: if particular groups of respondents tend to reply to a range of questions in a similar way, this overall grouping will often come out only weakly. If you try to follow the group of individuals who appear together in one corner cell of the first cross-tab, you can’t easily track whether they stay together in a cross-tab of other variables. Another type of approach may be more constructive: to derive synthetic variables – indicators – which bring together inputs from a range of questions, say into a measure of ‘compliance’, and to analyse those, by cross-tabulation or other methods. See section 2.8 below. If we have an analysis dataset with a row for each respondent and a column for each question, the derivation of a synthetic variable just corresponds to adding an extra column to the dataset. This is then used in analysis just like any other column. A profile for an individual will often comprise a set of values of a suite of indicators. 2.7 Looking for Respondent Groups Profiling is often concerned with acknowledging that respondents are not just a homogeneous mass, and distinguishing between different groups of respondents. Cluster analysis is a data-driven statistical technique that can draw out – and thence characterise – groups of respondents whose response profiles are similar to one another. The response profiles may serve to differentiate one group from another if they are somewhat distinct. This might be needed if the aim were, say, to define 22  © SSC 2001 – Approaches to the Analysis of Survey Data target groups for distinct safety net interventions. The analysis could help clarify the distinguishing features of the groups, their sizes, their distinctness or otherwise, and so on. Unfortunately there is no guarantee that groupings derived from data alone will make good sense in terms of profiling respondents. Cluster analysis does not characterise the groupings; you have to study each cluster to see what they have in common. Nor does it prove that they constitute suitable target groups for meaningful development interventions Cluster analysis is thus an exploratory technique, which may help to screen a large mass of data, and prompt more thoughtful analysis by raising questions such as:†¢ Is there any sign that the respondents do fall into clear-cut sub-groups? †¢ How many groups do there seem to be, and how important are their separations? †¢ If there are distinct groups, what sorts of responses do â€Å"typical† group members give? 2.8 Indicators Indicators are summary measures. Magazines provide many examples, e.g. an assessment of personal computers may give a score in numerical form like 7 out of 10 or a pictorial form of quality rating, e.g. Very good Good Moderate à  Poor Very Poor à ® This review of computers may give scores – indicators – for each of several characteristics, where the maximum score for each characteristic reflects its importance e.g. for one model:- build quality (7/10), screen quality (8/20), processor speed (18/30), hard disk capacity (17/20) and software provided (10/20). The maximum score over all characteristics in the summary indicator is in this case (10 + 20 + 30 + 20 + 20) = 100, so the total score for each computer is a percentage e.g. above (7 + 8 + 18 + 17 + 10) = 60%. The popularity of such summaries demonstrates that readers find them accessible, convenient and to a degree useful. This is either because there is little time to absorb detailed information, or because the indicators provide a baseline from which to weigh up the finer points. Many disciplines of course are awash with suggested indicators from simple averages to housing quality measures, social capital assessment tools, or quality-adjusted years of life. Of course new indicators should be developed only if others do nor exist or are unsatisfactory. Well-understood, well-validated indicators, relevant to the situation in hand are quicker and more cost-effective to use. Defining an economical set of meaningful indicators before data collection ought ideally to imply that at  © SSC 2001 – Approaches to the Analysis of Survey Data 23 analysis, their calculation follows a pre-defined path, and the values are readily interpreted and used. Is it legitimate to create new indicators after data collection and during analysis? This is to be expected in genuine ‘research’ where fieldwork approaches allow new ideas to come forward e.g. if new lines of questioning have been used, or if survey findings take the researchers into areas not  well covered by existing indicators. A study relatively early on in a research cycle, e.g. a baseline survey, can fall into this category. Usually this means the available time and data are not quite what one would desire in order to ensure well-understood, well-validated indicators emerge in final form from the analysis. Since the problem does arise, how does the analyst best face up to it? It is important not to create unnecessary confusion. An indicator should synthesise information and serve to represent a reasonable measure of some issue or concept. The concept should have an agreed name so that users can discuss it meaningfully e.g. ‘compliance’ or ‘vulnerability to flooding’. A specific meaning is attached to the name, so it is important to realise that the jargon thus created needs careful explanation to ‘outsiders’. Consultation or brainstorming leading to a consensus is often desirable when new indicators are created. Indicators created ‘on the fly’ by analysts as the work is rushed to a conclusion are prone to suffer from their hasty introduction, then to lead to misinterpretation, often over-interpretation, by enthusiast would-be users. It is all too easy for a little information about a small part of the issue to be taken as ‘the’ answer to ‘the problem’! As far as possible, creating indicators during analysis should follow the same lines as when the process is done a priori i.e. (i) deciding on the facets which need to be included to give a good feel for the concept, (ii) tying these to the questions or observations needed to measure these facets, (iii) ensuring balanced coverage, so that the right input comes from each facet, (iv) working out how to combine the information gathered into a synthesis which everyone agrees is sensible. These are all parts of ensuring face (or content) validity as in the next section. Usually this should be done in a simple enough way that the user community are all comfortable with the definitions of what is measured. There is some advantage in creating indicators when datasets are already available. You can look at how well the indicators serve to describe the relevant issues and groups, and select the most effective ones. Some analysts rely too much on data reduction techniques such as factor analysis or cluster analysis as a substitute for thinking hard about the issues. We argue that an intellectual process of indicator development should build on, or dispense with, more data-driven approaches. 24  © SSC 2001 – Approaches to the Analysis of Survey Data Principal component analysis is data-driven, but readily provides weighted averages. These should be seen as no more than a foundation for useful forms of indicator. 2.9 Validity The basic question behind the concept of validity is whether an indicator measures what we say or believe it does. This may be quite a basic question if the subject matter of the indicator is visible and readily understood, but the practicalities can be more complex in mundane, but sensitive, areas such as measurement of household income. Where we consider issues such as the value attached to indigenous knowledge the question can become very complex. Numerous variations on the validity theme are discussed extensively in social science research methodology literature. Validity takes us into issues of what different people understand words to mean, during the development of the indicator and its use. It is good practice to try a variety of approaches with a wide range of relevant people, and carefully compare the interpretations, behaviours and attitudes revealed, to make sure there are no major discrepancies of understanding. The processes of comparison and reflection, then the redevelopment of definitions, approaches and research instruments, may all be encompassed in what is sometimes called triangulation – using the results of different approaches to synthesise robust, clear, and easily interpreted results. Survey instrument or indicator validity is a discussion topic, not a statistical measure, but two themes with which statistical survey analysts regularly need to engage are the following. Content (or face) validity looks at the extent to which the questions in a survey, and the weights the results are given in a set of indicators, serve to cover in a balanced way the important facets of the notion the indicator is supposed to represent. Criterion validity can look at how the observed values of the indicator tie up with something readily  measurable that they should relate to. Its aim is to validate a new indicator by reference to something better established, e.g. to validate a prediction retrospectively against the actual outcome. If we measure an indicator of ‘intention to participate’ or ‘likelihood of participating’ beforehand, then for the same individuals later ascertain whether they did participate, we can check the accuracy of the stated intentions, and hence the degree of reliance that can in future be placed on the indicator. As a statistical exercise, criterion validation has to be done through sensible analyses of good-quality data. If the reason for developing the indicator is that there is no satisfactory way of establishing a criterion measure, criterion validity is not a sensible approach.  © SSC 2001 – Approaches to the Analysis of Survey Data 25 2.10 Summary In this guide we have outlined general features of survey analysis that have wide application to data collected from many sources and with a range of different objectives. Many readers of this guide should be able to use its suggestions unaided. We have pointed out ideas and methods which do not in any way depend on the analyst knowing modern or complicated statistical methods, or having access to specialised or expensive computing resources. The emphasis has been on the importance of preparing the appropriate tables to summarise the information. This is not to belittle the importance of graphical display, but that is at the presentation stage, and the tables provide the information for the graphs. Often key tables will be in the text, with larger, less important tables in Appendices. Often a pilot study will have indicated the most important tables to be produced initially. What then takes time is to decide on exactly the right tables. There are three main issues. The first is to decide on what is to be tabulated, and we have considered tables involving either individual questions or indicators. The second is the complexity of table that is  required – one-way, two-way or higher. The final issue is the numbers that will be presented. Often they will be percentages, but deciding on the most informative base, i.e. what is 100% is also important. 2.11 Next Steps We have mentioned the role of more sophisticated methods. Cluster analysis may be useful to indicate groups of respondents and principal components to identify datadriven indicators. Examples of both methods are in our Modern Methods of Analysis guide where we emphasise, as here, that their role is usually exploratory. When used, they should normally be at the start of the analysis, and are primarily to assist the researcher, rather than as presentations for the reader. Inferential methods are also described in the Modern Methods guide. For surveys, they cannot be as simple as in most courses on statistics, because the data are usually at multiple levels and with unequal numbers at each subdivision of the data. The most important methods are log-linear and logistic models and the newer multilevel modelling. These methods can support the analysts’ decisions on the complexity of tables to produce. Both the more complex methods and those in this guide are equally applicable to cross-sectional surveys, such as baseline studies, and longitudinal surveys. The latter are often needed for impact assessment. Details of the design and analysis of baseline surveys and those specifically for impact assessment must await another guide! 26  © SSC 2001 – Approaches to the Analysis of Survey Data  © SSC 2001 – Approaches to the Analysis of Survey Data 27 The Statistical Services Centre is attached to the Department of Applied Statistics at The University of Reading, UK, and undertakes training and consultancy work on a non-profit-making basis for clients outside the University. These statistical guides were originally written as part of a contract with DFID to give guidance to research and support staff working on DFID Natural Resources projects. The available titles are listed below. †¢ Statistical Guidelines for Natural Resources Projects †¢ On-Farm Trials – Some Biometric Guidelines †¢ Data Management Guidelines for Experimental Projects †¢ Guidelines for Planning Effective Surveys †¢ Project Data Archiving – Lessons from a Case Study †¢ Informative Presentation of Tables, Graphs and Statistics †¢ Concepts Underlying the Design of Experiments †¢ One Animal per Farm? †¢ Disciplined Use of Spreadsheets for Data Entry †¢ The Role of a Database Package for Research Projects †¢ Excel for Statistics: Tips and Warnings †¢ The Statistical Background to ANOVA †¢ Moving on from MSTAT (to Genstat) †¢ Some Basic Ideas of Sampling †¢ Modern Methods of Analysis †¢ Confidence Significance: Key Concepts of Inferential Statistics †¢ Modern Approaches to the Analysis of Experimental Data †¢ Approaches to the Analysis of Survey Data †¢ Mixed Models and Multilevel Data Structures in Agriculture The guides are available in both printed and computer-readable form. For copies or for further information about the SSC, please use the contact details given below. Statistical Services Centre, The University of Reading P.O. Box 240, Reading, RG6 6FN United Kingdom tel: SSC Administration +44 118 931 8025 fax: +44 118 975 3169 e-mail: [emailprotected] web: http://www.reading.ac.uk/ssc/

Theories and studies about reducing racial prejudice

Theories and studies about reducing racial prejudice Everyone has a race or ethnic group that they see themselves as being part of. On the other hand, not all people are exposed to stinging words or physical harm from a prejudiced individual because of the color of their skin. According to the Bureau of Justice Statistics (2005), there are approximately 210,000 hate crimes a year; racial prejudice is the motivation for over half of them. This paper will discuss theories and studies on ways to reduce racial prejudice. Racial prejudice has been around since groups of people could distinguish themselves from one another (Milner, 1983). The 1920s were when prejudice started catching the attention of psychologists as a social phenomenon that needed to be studied (Duckitt, 1992). Samelson (1978) talked about how tests between races were first meant to measure individuality but soon the authors were publishing empirical evidence that Whites were superior to Blacks (as cited in Duckitt, 1992, p. 1185). Milner (1983) states that prejudice occurs because people become frustrated, need a scapegoat, or because they are feeling some anxiety and need a way to release it. One of the first texts on prejudice and reducing prejudice was written by Gordon W. Allport. Encouraged by Robin Williams study on conditions that further the reduction of racism, Allport wrote The Nature of Prejudice in which he discusses his contact hypothesis (Utsey, Ponterotto, Porter, 2008). Allport (1954) stated that prejudice may be reduced by equal status contact between majority and minority groups in the pursuit of common goals (p. 281). Allport (1954) also says there are eight different areas of contact, causal, residential, occupational, recreational, religious, civic and fraternal, political, and goodwill intergroup activities. Allport (1954) says that both state and federal legislation can be used to pass antidiscrimination laws and have public agencies enforce these laws. According to Allport (1954), there are six programs that can be used to reduce prejudice. They are formal educational methods, contact and acquaintance programs, group retraining methods, mass media, exhortation, and individual therapy. Allport (1954) feels that individual therapy is the best one, yet no study has been convincing of this. Allport (1954) describes formal educational methods as teaching about prejudice in the school setting. There are five types of formal educational methods. There is the informational approach, direct approach, indirect approach, the approach through vicarious experience, and the project method. The next method is contact and acquaintance programs which means that White people and Black people get together and get to know each other. The third method is group retraining. In group retraining, the outgroup members and the ingroup members switch roles and try to become empathetic to each other. The fourth method is mass media in which messages are sent in the media spreading information on tolerance of others. The next method is exhortation which is like religion in which leaders spread the message of tolerance of other people. Finally, there is individual therapy in which a person meets with a therapist to change their way of thinking. Blincoe and Harris (2009) talk about three major programs that cause a minimization in racial prejudice. Cooperation is similar to Allports (1954) contact theory. This program has been used in jigsaw classrooms in which children are broken up into racially varied group and then each child is given a piece of information to teach to the others (Aronson Bridgeman, 2007; Blincoe Harris, 2009). Along with the children learning information, they also showed higher self-esteem, liked school more, and for minorities, their school grades had improved (Aronson Bridgeman, 2007). The tolerance program is synonymous with political tolerance and the respect program is supposed to reinforce and encourage diversity (Blincoe Harris, 2009). Crisp and Turner (2009) hypothesize that imagining contact with an outgroup can have a close or same effect on diminishing prejudice as actual contact with an outgroup. Turner, Crisp, and Lambert, (2007) found that participants who imagined an optimistic interaction with an outgroup member conveyed more positive attitudes and less prejudice than those who did not (as cited in Crisp Turner, 2009). Stathi and Crisp (2008) did a study that showed that even though projection of positive self traits is higher for ingroups than outgroups (Clement Krueger, 2002), positive imagined contact leads to greater projection of positive traits to outgroups (as cited in Crisp Turner, 2009, p. 234). In addition to contact theory, there is the goal based approach which consists of three goals that people need to reach. These are comprehension goals, self-enhancement goals, and motivation to avoid prejudice (Kunda Spencer, 2003). Kunda and Spencer (2003) say that comprehension goals include the need to understand events, reduce the complexity of the environment, gain cognitive clarity, and form rational impressions. Stereotypes serve these needs by enabling perceivers to simplify and understand the huge amounts of social information that they confront and to make inferences that go beyond available information (p. 524). They also say that self-enhancement goals include the need to protect and enhance self-esteem. Lastly, motivation to avoid prejudice inhibits the activation of stereotypes. When people notice that they are treating others differently because of their skin color, they will feel the discrepancies because they know it is not right. Therefore they feel guilty which ma kes them repress their prejudiced thoughts. Another part of the goal-based theory is why there is prejudice in the first place. Kenrick, Neugberg, and Cialdini (2009) feel that there are two things that prejudice does for people, it helps us gain economic resources and the characteristics of the other groups bring our economic goals to our attention. The first way to achieve the goals of the goal-based theory is to attempt to change the character of the prejudiced person. The second is to change the situation in which the prejudiced person feels like they can discriminate against others. Next is to give people a different way to satisfy their goals and last is to activate goals incompatible with prejudice, stereotyping, and discrimination. Part of the goal-based approach involves looking at the point of view of other people. Galinksy and Moskowitz (2000) say that when a person looks at themselves, they have higher favorable responses to the ingroup. Turner (1987) says favoritism increases toward the in-group (as cited in, Galinsky Moskowitz, 2000, p. 709). Therefore, thinking that you are part of the outgroup will increase positive responses to them and decrease prejudice thoughts about them (Galinsky Moskowitz, 2000). In the judicial area, the goal-based approach has some significance. Studying about prejudice and ingroups and outgroups can become very important especially for those who may be suing another person in civil court for injury that they could have caused. People tend to like others who are similar to them (Kerr, Hymes, Anderson, Weathers, 1995). If a juror feels that they are not similar to the plaintiff in a malpractice case that juror might feel that the plaintiff should get a lower amount of money to compensate for the damages or perhaps believe they should not get any money at all. The same applies to the defendant. If the juror feels they are similar to the defendant than they could be more lenient on his punishment (Green Bornstein, 2003). However, Marques and Yzerbyt (1988) say that the opposite effect can also happen. That is, the jurors are harsher on an ingroup member because they are part of the ingroup and they pose a threat to the positive image of the ingroup members. They call it the black sheep effect in which positively viewed ingroup members are viewed better than outgroup members but negatively viewed ingroup members are viewed as being worse than outgroup members (as cited in, Green Bornstein 2003). Finally, there is the ignorance hypothesis. People experience prejudiced thoughts because they simply do not know any better (Kenrick, Neugberg, Cialdini, 2009). If everyone would interact with the other groups, they wouldnt stereotype individuals of other groups. However, Stephan and Stephan (1996) say research shows that this approach does little to reduce prejudice (as cited in, Kenrick, Neuburg, Cialdini, 2009). Case (2007) did a study in which college students were required to take a course on diversity. The course was designed to heighten recognition of White privilege and racism, raise support for affirmative action, and decrease prejudice, guilt, and fear of other races. The students took a survey at the beginning of the course which measured White privilege, awareness of racism, and the students level of racism to different ethnic groups. The same survey was given at the end of the semester as well. Results showed white privilege, awareness of racism and support for affirmative action increased. However, students reported greater fear of other races. Students levels of racism remained constant except for racism against Latinos, which increased. Case explains this as possibly being by chance or that the course could have actually increased prejudice. Blanchard, Lilly, and Vaughn (1991) hypothesized that hearing another person express strong antiracist opinions would have more of an effect than hearing another person express equal opinions or opinions that were more accepting of racism. They also speculated that when a person hears another person express strong support of racism, the first person showed less support of antiracism. They did two studies in which they interviewed college students in a group with a confederate who either openly expressed strong antiracist views or strong racist views when asked about a false situation in what should happen to another student who wrote racist notes. There was a neutral condition in which the participant answered the questions first and in the other condition the confederate answered first. The authors hypotheses were confirmed in both of the experiments. In 2007, four studies were done by Turner, Hewstone, and Loci that investigated self-disclosure as a mediator of the effect of cross-group friendship and vicarious experiences of such friendship (p. 371). The subjects for studies one were children between the ages of eight and twelve. The ages of the children for subjects two and three were 12 through 16. The last study included undergraduate students. In Study One, students were first given tasks that required them to categorize photographs of faces as negative or positive and White or Asian. The last tasks were to categorize White/Positive or Asian/Negative and White/Negative or Asian/Positive. In Study Two, the students were given questionnaires on their thoughts on the other ethnic group. The third study was the same as the second study except the experimenters used a larger group. In Study Four, White participants were given a questionnaire that measured predictor variables, mediator variables, and explicit outgroup attitude. So me of questions were, How often do you discuss intimate or personal issues with people who are Asian? (p. 380) and How rewarding are the interactions you have with Asian people? (p. 380). All four studies found that self-disclosure positively predicted explicit outgroup attitude. Vrij, Akehurst, and Smith (2003) conducted a study where people were shown cue cards and then were given surveys to measure prejudice. They focused on seven principles, that they say decrease prejudice when used in public campaigns. The seven principles are (1) an emphasis on similarities; (2) positive similarities in a positive context; (3) many representative members; (4) provision of explicit information; (5) employ a credible source; (6) state illegality; (7) central and peripheral routes to persuasion (p. 285). Each of the cue cards had one of the seven principles or the opposite of it. For example, state illegality was shown on one cue card as one White man and one Black man approximately the same age with wording above them that said These two men applied for a job as an Accounts Manager. The man on the left was turned down because he is Black (p. 291); the other card was the same as the first one but had the wording IT IS ILLEGAL TO DISCRIMINATE ON THE GROUNDS OF RACE (RACE R ELATIONS ACT, 1976) (p. 291). Subjects were then given a survey that measured their prejudice. Vrij, Akehurst, and Smith found that if the subject viewed a card that did not have one of the seven principles, their prejudice had increased versus if they had seen one of the principles. The two principles that had the most effect were emphasis of similarities and similarities in a positive context. Carpenter, Za ´rate, and Garzas study that was done in 2007, focused on using differences and individuality to reduce prejudice in groups that are African American, White American, Mexican American, and Mexican National. In Experiment One, the African American and White American participants were first primed with stories that had an emphasis on the personal self or others. Then, they filled out questionnaires while looking at pictures of African Americans and White Americans. The White Americans, who were primed to have an emphasis on others, had reduced prejudice. However, the African Americans showed no difference in prejudice levels. In Experiment Two, White Americans, Mexican Americans, and Mexican Nationals took self-esteem tests and then answered questions on all three groups. Carpenter, Za ´rate, and Garza (2007) found that self-esteem did not have any effect on prejudice and that looking at ways that your own ethnic group is different from other groups can lessen prejudi ce. Pettigrew and Tropp (2006) did a meta-analysis of over 500 studies and 713 independent samples that tested the intergroup contact theory. Their findings showed that intergroup contact does decrease intergroup prejudice. Pettigrew and Tropp go on to say that the conditions are not independent but entwined with each other. They also feel that intergroup contact can be utilized to end prejudice against other underrepresented groups. Racism not only exists among individual people but also in government forms. Billingsley and Giovannoni (1972) have been doing studies that show that African American children have been consistently counted out from services provided by child welfare establishments, they believe this is due to racism that exists in these institutions (as cited in, Miller Ward, 2008). Miller and Ward (2008) say there has been overrepresentation of African Americans in the welfare system for a long time. They then go on to talk about the Breakthrough Series Collaborative (BSC) methodology was used to analyze the welfare systems racism and then identify strategies to reduce the racial disproportions. The BSC theory of change has six areas which are (1) increase the awareness and understanding of the issue, (2) identify challenges and test strategies for improvement, (3) implement site-level policy and practice improvements, (4) spread the improvements throughout the larger system, (5) sustain system-wi de improvements, and (6) improve child and family outcomes (p. 227). Many participants of the program reported being able to achieve a fully functional program in their location. However, the participants said they had difficulty spreading changes from their location to a larger system. More work still needs to be done to stop the racial prejudice that occurs in the welfare system. There are also racial discrepancies in the health care system. For example, according to the Centers for Disease Control, in 2006, the age-adjusted death rate for White Americans of both sexes, was 764.4 and for African Americans of both sexes, it was 982.0 (Heron, Hoyert, Murphy, Xu, Kochanek, Tejada-Vera, 2009). In 2002, Dovidio et al. conducted a study of racism that occurs during an emergency. White subjects were half as likely to help a Black person as they were to help a White person. While the participants opposed that the idea they were racist, it was the only difference in the fabricated emergency (as cited in Carlson Chamberlain, 2004, p. 375). Carlson and Chamberlain (2004) say that to reduce the health disparities between White Americans and African Americans, there must be a change in the research areas that combine social conditions with the physiological pathways to health and disease and that we need to join together on emotional levels to understand each other to c hange racial attitudes. In addition to healthcare and welfare, racism has even showed up in our grocery stores. In a study that was done in 2003, Topolski, Boyd-Bowman, and Ferguson found differences in the quality of fruits in grocery stores that were part of the same chain but were located in different parts of the city. They collected fruit samples from stores that were located in neighborhoods that had high socio-economic status and low socio-economic status. More minorities lived in the lower income neighborhoods. The quality of the fruit in the high SES neighborhood was better than the fruit that came from the other neighborhood, as judged by students who examined, ate the fruit and then rated them. As you can see, there is hope for eliminating racial prejudice. On the other hand, a lot of the studies that I presented in this paper were done with children and college students. It still leaves out a majority of the population. Yet, I think we have come a long way from previous generations in accepting others, especially in the case of race and ethnicity. If we eliminate or even reduce racial prejudice, then minority children will do better in school, they will have better economic and career opportunities, and will experience lower rates of crime against them. With the current research on racial prejudice, we could also apply these theories to sexism, homophobia, and ageism. Hopefully, in time, Andy Warhols I think everybody should like everybody quote will finally be true.