the “Test Yourself” section on p. 199 in Ch. 9 of . your response with your classmates. the “Test Yourself” section on p. 207 in Ch. 9 of Exploring Psychology. your response with your classmates. what you have learned about the Literature Review process, this week. to one or more of the following prompts in one to two paragraphs 1. Provide citation and reference to the material(s) you discuss. Describe what you found interesting regarding this topic, and why. 2. Describe how you will apply that learning in your daily life, including your work life. 3. Describe what may be unclear to you, and what you would like to learn. Using Figure 1.2 in Ch. 1 of , a flowchart using Microsoft® Word or a similar program that helps you identify what research design to use for your research question. In some ways, your work on the first eight chapters of has been done to prepare you for the next four, all of which deal with particular types of research designs or research methods. In this chapter, you will learn about nonexperimental research methods, which are ways of looking at research questions without the direct manipulation of a variable. discusses another nonexperimental approach: qualitative methods. Why a separate chapter? Because the whole area of qualitative methods stands alone as a somewhat unique approach to asking and answering social and behavioral science research questions.So, let’s turn our attention to the techniques we will deal with here.For example, if you wanted to understand the factors that may be related to why certain undergraduates smoke and why others do not, you might want to complete some type of survey, one of the descriptive techniques that will be covered in this chapter. Or, if you were interested in better understanding the relationship between risk-taking behavior and drug abuse, perhaps the first (but not the last) step would be to conduct a correlational study in which you would learn about questions of a correlational nature. You would be examining the association between variables and learning about the important distinction between association (two things being related since they share something in common) and causality (one thing causing another).This chapter focuses on descriptive research questions, how they are asked and how they are answered. It’s the first chapter on methods before we move on to qualitative, true experimental, and quasi-experimental methods. Although several factors distinguish different types of research from one another, probably the most important factor is the type of question that you want to answer (see the summary chart on page 00 in ). If you are conducting descriptive research, you are trying to understand events that are occurring in the present and how they might relate to other factors. You generate questions and hypotheses, collect data, and continue as if you were conducting any type of research.Descriptive research describes the current state of some phenomenon.The purpose of descriptive research is to describe the current state of affairs at the time of the study. For example, if you want to know how many teachers use a particular teaching method, you could ask a group of students to complete a questionnaire, thereby measuring the outcome as it occurs. If you wanted to know whether there were differences in the frequency of use of particular types of words among 3-, 5-, and 7-year-olds, you would describe those differences within a descriptive or developmental framework.The most significant difference between descriptive research and causal comparative or experimental research (discussed in detail in ) is that descriptive research does not include a treatment or a control group. You are not trying to test the influence of any variable upon another. In other words, all you are doing for readers of your research is painting a picture. When people read a report that includes one of the several descriptive methods that will be discussed, they should be able to envision the larger picture of what occurred. There may be room to discuss why it occurred, but that question is usually left to a more experimental approach.Although there are many different types of descriptive research, the focus of this discussion will be on survey research, and correlational studies in which relationships between variables are described. The best application of sampling in theory and practice can probably be found in survey research. Survey researchers attempt to study directly the characteristics of populations through the use of surveys. You may be most familiar with the types of surveys done around election time, wherein relatively small samples of potential voters (about 1,200) are questioned about their voting intentions. To the credit of the survey designers, the results are often very close to the actual outcomes following the election. , also called sample surveys, examines the frequency and relationships between psychological and sociological variables and taps into constructs such as attitudes, beliefs, prejudices, preferences, and opinions. For example, a sample survey could be used to assess the following: The basic tool used in survey research is the . Interviews (or oral questionnaires) can take the form of the most informal question-and-answer session on the street to a highly structured, detailed interaction between interviewer and interviewee. In fact, many of the points that were listed for questionnaires also apply to interviews. For example, although you need not be concerned about the physical format of the questions in an interview (because the respondent never sees them), you do need to address such issues as transitioning between sections, being sensitive to the type of information you are requesting, and being objective and straightforward.Interviews are much more challenging and difficult to do well than just discussing a topic with someone.Most interviews begin with what is called , or neutral information, about the respondent such as age, living arrangements, number of children, income, gender, and educational level. Such information helps the interviewer accomplish several things.First, it helps establish a rapport between the interviewer and the interviewee. Such questions as “Where did you go to college?” or “How many children do you have?” are relatively nonthreatening.Second, it establishes a set of data that characterizes the person being interviewed. These data can prove invaluable in the analysis of the main focus of the interview which comes later on in the survey.Interviews contain two general types of questions: structured and unstructured questions. or questions have a clear and apparent focus and call for an explicit answer. They are comprehensible to the interviewer as to the interviewee. Such questions as “At what age did you start smoking?” and “How many times have you visited this store?” call for explicit answers. On the other hand, or questions allow the interviewee to elaborate upon responses. Such questions as “Why were you opposed to the first Persian Gulf War?” or “How would you address the issue of teenage pregnancy?” allow for a more broad response by the interviewee. In both cases, the interviewer can follow up with additional questions.Interviews can be especially helpful if you want to obtain information that might otherwise be difficult to come by, including firsthand knowledge of people’s feelings and perceptions. For example, in a study conducted by M. L. Smith and L. A. Shepard ( ), interviews with teachers and parents were part of a multifaceted approach to understanding kindergarten readiness and retention. In this study, interviewing was combined with other techniques such as in-class observations and the analysis of important documents. These researchers put the interview results to good use when they examined these outcomes in light of other information they collected throughout the study.On the positive side, interviews offer great flexibility by letting you pursue any direction (within the scope of the project) with the questions. You could also note the interviewee’s nonverbal behavior, the setting, and other information that might provide valuable information. Another advantage of interviews is that you can set the general tone and agenda at your own convenience (to a point, of course).There is also a downside to interviews. They take time, and time is expensive. Interviewing 10 people could take 20–30 hours including travel time and such. Also, because interviews have less anonymity than, for example, a questionnaire, respondents might be reluctant to come forward as honestly as they might otherwise. Other disadvantages are your own biases and the lack of a standardized set of questions. A good interviewer will probe deeply for additional information, perhaps of a different type, than would another interviewer who started with the same questions. Asking follow-up questions is an excellent practice, but what do you do about the interview where probing did not lead to the same information and thus produced different results? What do you think a primary advantage of an interview is over a more structured tool such as a questionnaire, and when might you want to use the interview technique? The development of an interview begins much like that for any proposal for a research project. Your first step is to state the purpose of the interview by taking into account your goals for the project. Then, as before, you review the relevant literature to find out what has been done in the past and whether other interview studies have been conducted. You may even find an actual interview that was previously used and be able to use parts of that in your own research. This is a very common practice when researchers use the same interview, say, 10 years later to look for changes in trends.Second, select a sample that is appropriate for your study, both in characteristics and in size. If you want to know about feelings regarding racial unrest, you cannot question only white citizens—you need to address all minorities. Similarly, even if interviews take lots of time and effort, you cannot skimp on sample size with the thought that what is lost in sample size can be made up in richness and detail. It does not work that way.Next, the interview questions need to be developed. As you know by now, questions, whether structured or unstructured, need to be clear and concise without any hidden agenda, double negatives, 75-cent words that cannot be understood, and so forth. One of the best ways to determine the appropriateness of your interview is by field-testing it. Use it with people who have the same characteristics as the intended audience. Listen to their feedback and make whatever changes you find necessary.After the interview form is (more or less) finished, it is time to train the interviewers. Most of the traits you want in an interviewer are obvious: They should be polite, neatly dressed, uncontroversial in appearance, and responsible enough to get to the interview site on time. These qualities, however, are not enough. Interviewers must learn how to go beyond the question should the need arise. For example, if you are asking questions about racial discrimination, the respondent might mention, “Yes, I sometimes feel as if I am being discriminated against.” For you not to ask “Why?” and to follow up on the respondent’s answer would result in the loss of potentially valuable and interesting information. The best way to train is to have an experienced interviewer watch the trainees interview a practice respondent and then provide feedback.Finally, it is time to conduct the actual interviews. Allow plenty of time, and go to it. Do not be shy, but do not be too aggressive either. If you have worked hard at getting ready for the interview, you should not encounter any major problems. Nonetheless, there are certain things you should keep in mind to make your interview run a bit more smoothly and be more useful later, when it comes time to examine the results of your efforts.No one is perfect, but you should strive to adhere to these 10 guidelines about interviewing as well as you can.With that in mind, here are the 10 commandments of interviewing (drumroll, please). Keep in mind that many, if not all of these, could also be classified as interviewer effects, in which the behavior of the interviewer can significantly affect the outcome. Have you ever been at home during the dinner hour and the phone rings, and the person on the other end of the line wants to know how often you ride the bus, recycle your newspaper, use a computer, or rent a car?Those calls represent one of several types of survey research, all of which are descriptive in nature. In addition to interviews—the primary survey research method—and telephone surveys, surveys include panels or focus groups (in which a small group of respondents is interviewed and reinterviewed) and mail questionnaires. Survey research starts out with a general plan (a ) of what activities will occur when. The plan begins with the objective of the study, leads into the various methods that may be used to collect the data, and finishes with a final report and a summary of the findings. Some type of analysis of the frequencies of these responses can be performed to answer the question about parents’ attitudes toward punishment. Collecting survey data is hard work. It means constantly seeking subjects and dealing with lots of extraneous sources of variance that are difficult to control. It is somewhat of a surprise, however, how relatively easy it is to establish the validity of such data. For example, one way to establish the validity of the data gained from an interview is to seek an alternative source for confirmation. Public records are easy to check to confirm such facts as age and party affiliation. Respondents can even be interviewed again to confirm the veracity of what they said the first time. There is no reason why people could not lie twice, but a good researcher is aware of that possibility and tries to confirm factual information that might be important to the study’s purpose. 1214152332 461314 7 6 Like all other research methods, survey research has its ups and downs. Here are some ups. First, survey research allows the researcher to get a very broad picture of whatever is being studied. If sampling is done properly, it is not hard to generalize to millions of people, as is done on a regular basis with campaign polling and such. Along with such powers to generalize comes a big savings in money and time.Second, survey research is efficient in that the data collection part of the study is finished after one contact is made with respondents and the information is collected. Also, minimal facilities are required. In some cases, just a clipboard and a questionnaire is enough to collect data.Third, if done properly and with minimal sampling error, surveys can yield remarkably accurate results.The downs can be serious. Most important are sources of bias which can arise during interviews and questionnaires. occurs when the interviewer subtly biases the respondent to respond in one way or another. This bias might take place, for example, if the interviewer encourages (even in the most inadvertent fashion) approval or disapproval of a response by a smile, a frown, looking away, or some other action.On the other hand, the interviewee might respond with a bias because he or she may not want to give anything other than socially acceptable responses. After all, how many people would respond with a definite “yes!” to the question, “Do you beat your spouse?”These threats of bias must be guarded against by carefully training interviewers to be objective and by ensuring that the questions neither lead nor put respondents in a position where few alternatives are open.Another problem with survey research is that people may not respond, as in the case of a mail survey. Is this a big deal? It sure can be. Nonresponders might constitute a qualitatively distinct group from responders. Therefore, findings based on nonresponders will be different than if the entire group had been considered. The rule? Go back and try to get those who didn’t respond the first time. You read about ethics and some guidelines in . What might be some conflicts tha
t can arise with those ethical principles and the use of the various survey methods we discussed earlier? describes the linear relationship between two or more variables without any hint of attributing the effect of one variable on another. As a descriptive technique, it is very powerful because this method indicates whether variables (such as number of hours of studying and test score) share something in common with each other. If they do, the two are correlated (or co-related) with one another.In , the correlation coefficient was used to estimate the reliability of a test. The same statistic is used here, again in a descriptive role. For example, correlations are used as the standard measure to assess the relationship between degree of family relatedness (e.g., twins, cousins, unrelated) and similarity of intelligence test scores. The higher the correlation, the higher the degree of relatedness. In such a case, you would expect that twins who are raised in the same home would have more similar IQ scores (they share more in common) than twins raised in different homes. And they do! Twins reared apart share only the same genetic endowment, whereas twins (whether monozygotic [one egg] or dizygotic [two eggs]) reared in the same home share both hereditary and environmental backgrounds. The most frequent measure used to assess degree of relatedness is the correlation coefficient, which is a numerical index that reflects the relationship between two variables. It is expressed as a number between 21.00 and 11.00, and it increases in strength as the amount of variance that one variable shares with another increases. That is, the more two things have in common (like identical twins), the more strongly related they will be to each other (which only makes sense). If you share common interests with someone, it is more likely that your activities will be related than if you compared yourself with someone with whom you have nothing in common.For example, you are more likely to find a stronger relationship between scores on a manual dexterity test and a test of eye–hand coordination than between a manual dexterity test and a person’s height. Similarly, you would expect the correlation between reading and mathematics scores to be stronger than that between reading and physical strength. This is because performances on reading and math tests share something in common with each other (intellectual and problem-solving skills, for example) than a reading test and, say, weight-lifting performance.Correlations can be direct or positive, meaning that as one variable changes in value, the other changes in the same direction, such as the relationship between the number of hours you study and your grade on an exam. Generally, the more you study, the better your grade will be. Likewise, the less you study, the worse your grade will be. Notice that the word “positive” is sometimes interpreted as being synonymous with “good.” Not so here. For example, there is a negative correlation between the amount of time parents spend with their children and the child’s level of involvement with juvenile authorities. Bad? Not at all.Positive correlations are not “good” and negative ones are not “bad.” Positive and negative have to do with the direction of the relationship and nothing else.Correlations can also reflect an indirect or negative relationship, meaning that as one variable changes in value in one direction, the other changes in the opposite direction, such as the relationship between the speed at which you go through multiple-choice items and your score on the test. Generally, the faster you go, the lower your score; the slower you go, the higher your score. Do not interpret this to mean that if you slow down, you will be smarter. Things do not work like that, which further exemplifies why correlations are not causal. What it means is that, for a specific set of students, there is a negative correlation between test-taking time and total score. Because it is a group statistic, it is difficult to conclude anything about individual performance and impossible to attribute causality.The two types of correlations we just discussed are summarized in .Interestingly, the important quality of a correlation coefficient is not its sign, but its A correlation of 2.78 is stronger than a correlation of 1.68, just as a correlation of 1.56 is weaker than a correlation of 2.60. The most frequently used measure of relationships is the , represented by letter followed by symbols representing the variables being correlated. The symbol xyrepresents a correlation between the variables and To compute a correlation, you must have a pair of scores (such as a reading score and a math score) for each subject in the group with which you are working. For example, if you want to compute the correlation between the number of hours spent studying and test score, then you need to have a measure of the number of hours spent and a test score for each individual.The absolute value of the correlation coefficient, not the sign, is what’s important.As you just read, correlations can range between −1.00 and +1.00 and can take on any value between those two extremes. For example, look at , which shows four sets of data (A, B, C, and D) represented by an accompanying scattergram for each of the sets. Increases in valueIncreases in valuePositive or directThe taller one gets (X), the more one weighs (Y)Decreases in valueDecreases in valuePositive or directThe fewer mistakes one makes (X), the fewer hours of remedial work (Y) one participates inIncreases in valueDecreases in valueNegative or indirectThe better one behaves (X), the fewer in-class suspensions (Y) one hasDecreases in valueIncreases in valueNegative or indirectThe less time one spends studying (X), the more errors one makes on the test (Y)The scattergram is a visual representation of the correlation coefficient of the relationship between two variables.A is a plot of the scores in pairs. In set A, the correlation is +1.70. (You will see how to compute that value in a moment.)To draw a scattergram, follow these steps: Now look at data set B, where the correlation is only .32, which is substantially weaker than .70. You can see that the stronger correlation (set A) is characterized in the following ways: The data in set A show a high positive correlation (.70), whereas the data in set B show a much lower one (.32). The data in set C show a high negative correlation (− .82) and, just as with a high positive correlation, the coordinates that represent the intersection of two data points align themselves along a diagonal (in this case, from the upper left-hand corner to the lower right, approaching a 45° angle). The last data set, set D, shows very little relationship (− .15) between the and the variables, and the accompanying plot of the coordinates reveals a weak pattern. In other words, a line drawn through these points would be almost flat or horizontal. In summary, the stronger the formation of a pattern and the more the pattern aligns itself in a 45° angle (either from the lower left-hand corner of the graph to the upper right-hand for positive correlations, or from the upper left-hand corner of the graph to the lower right-hand corner for negative correlations), the stronger the visual evidence of the existence of a relationship between two variables. Correlations can be negative or positive, but give an example of how does not carry a pejorative meaning and outcomes are not always good. The easiest manual way to compute the correlation between two variables is through the use of the raw score method. The formula for xy (where xy represents the correlation between and ) is as follows:The Pearson correlation coefficient is the most frequently computed type of correlation.r xy = n Σ XY − Σ X Σ Y√ [ n Σ X 2 − ( Σ X ) 2 ] [ n Σ Y 2 − ( Σ Y ) 2 ] where Let’s look at a simple example where the correlation coefficient is computed from data set C shown in . The mean for variable is 6.3, and the mean for variable is 4.6. Here is what the finished equation looks like:r xy = − .373 [ 32.1 ] [ 62.4 ] = − .82Try it yourself and see if you can get the same result ( xy = −.82). You can also use SPSS or Excel to get the answer. Grade1.00.321.039Reading.3211.00.605Math.039.6051.00The correlation is the expression of the relationship between the variables of X and Y, represented as xy. What happens if you have more than two variables? Then you have more than one correlation coefficient. In general, if you have variables, then you will have “ taken two at a time” pairs of relationships. In , you can see a correlation matrix, or a table revealing the pairwise correlations between three variables (grade, reading score, and mathematics score). Each of the three correlation coefficients was computed by using the formula described earlier.You may notice that the diagonal of the matrix is filled with 1.00s because the correlation of anything with itself is always 1. Also, the coefficients to the right of the diagonal and to its left form a mirror image. The correlations for the other “half” of the matrix (above or below the diagonal of 1.00s in ) are the same. The correlation coefficient is an interesting index. It reflects the degree of relationship between variables, but it is relatively difficult to interpret as it stands. However, there are two ways to interpret these general indicators of relationships.To interpret the meaning of the correlation coefficient, look to the correlation of determination.The first method is the “eyeball” method, in which correlations of a certain value are associated with a certain nominal degree of relationship such that:Correlations betweenAre said to be.8 and 1.0Very strong.6 and .8Strong.4 and .6Moderate.2 and .4Weak.0 and .2Very weakRemember: Do not be fooled by these numbers. Even the weakest correlation (such as .1) can be statistically significant if the sample upon which it is based is large enough and sufficiently approaches the size of the population. You read about the significance versus meaningfulness distinction in . 0.10.01 0.20.04.1 to .230.30.09.2 to .350.40.16.3 to .470.50.25.4 to .590.60.36.5 to .6110.70.49.6 to .7130.80.64.7 to .8150.90.81.8 to .9171.01.00.9 to 1.019A sounder method for interpreting the correlation coefficient is to square its value and then compute the . This value, , is the amount of variance that is accounted for in one variable by the other. In other words, it allows you to estimate the amount of variance that can be accounted for in one variable by examining the amount of variance in another variable. Thus, if the correlation between two variables is .40, then the coefficient of determination is .16. Sixteen percent (16%) of the variance in one variable can be by the variance in the other variable; 84% (or 100%-16%) of the variance is unexplained. This portion of variance is often referred to as the . It is interesting to compare how the amount of variance explained in the relationship between two variables changes as the correlation gets stronger. The change isn’t as predictable as you might think. shows the simple correlation coefficient (the first column) and the coefficient of determination (the second column). Notice the change in the amount of variance accounted for as the value of the correlation coefficient increases. For example, if the correlation is increased from .4 to .5, the increase in the amount of variance accounted for is 9%. But if the correlation is increased a similar amount (say, from .6 to .7, which is still .1), then the increase in the amount of variance accounted for is 13%. The increase in the variance explained is not linear; therefore, the higher the correlation is, the larger the “jump” in explained variances. is a graphic illustration of what is shown in . As the correlation increases in value, an increasingly larger amount of variance is accounted for. That’s why the line shown in curves—the amount of variance ( ) increases disproportionately as the value of the correlation coefficient (the axis) increases and that’s why the higher the value of the correlation, the more relative variance you can explain as a relationship between variance than for a lower correlation value. Of the various research method tools you have learned about so far, what are some of the advantages and disadvantages of the correlational research methods? Is a nonexperimental—descriptive or correlational—design right for you? This is not really the question that should be asked. Rather, you should ask if your subject of interest demands that you use the tools suggested by the descriptive method. As emphasized before, the question that is asked determines the way it is answered. If you want to investigate how the Oklahoma settlers of the 1930s raised their children or how child rearing has changed, historiography may be for you. And what does the descriptive method offer? It provides an account of an event, often in such detail that it serves as a springboard for other questions to be asked and answered. Case studies, , and correlational studies describe a particular phenomenon in a way that communicates the overall picture of whatever is being studied. Although these methods do not allow the luxury of implying any cause-and-effect relationship between variables, their use provides the tools needed to answer questions that are otherwise unanswerable. .Write out several questions that would be interesting to study using survey research. Create a few questions of a survey nature for each of the studies. .Name two advantages and two disadvantages to interviews. .Write three potential follow-up questions to this initial interview question: What is your attitude toward eliminating score keeping in children’s sports? .Briefly outline the five steps of developing an interview. .Rank the following correlation coefficients in order of their strength from strongest to weakest. .What is wrong with the following argument? The relationship between the number of hours you spend studying is directly related to how well you do on school tests. Therefore, if you do not do well on a test, it means that you did not study long enough. .Indicate the type of correlation each of the following relationships describes: positive, negative, or no relationship. .For each of the three relationships in exercise #4, provide an example. .Tell whether the following hypotheses are correlational in nature. .Improve the following interview questions: .What is the purpose of descriptive research? .Provide an example of when descriptive research might be the appropriate method to use to answer your research question. And while you are at it, what is your question? .Which of the following statements about correlation coefficients are true? .What is an example of where a correlation might be significant but not meaningful? .Examine the relationship between consumption of milk during dinner and nighttime bedwetting and find a significant correlation of .25. How would you interpret the meaningfulness of this finding? .What does the coefficient of determination mean? What would the value of the coefficient of determination be for two variables with a correlation of .60? What would be the value of the coefficient of alienation? A huge number of educational databases (as part of the main reading room of the Library of Congress) to start your own descriptive research can be found at . You can find everything here from ERIC to a listing of universities worldwide. Dr. John Suler at Rider University gives tips on how to conduct the interview and how appropriately to include information from interviews in your research paper at . Walk down the hall in any building on your campus where social and behavioral science professors have their offices in such departments as psychology, education, nursing, sociology, and human development. Do you see any bearded, disheveled, white-coated men wearing rumpled pants and smoking pipes, hunched over their computers and mumbling to themselves? How about disheveled, white-coated women wearing rumpled skirts, smoking pipes, hunched over their computers, and mumbling to themselves?Researchers hard at work? No. Stereotypes of what scientists look like and do? Yes. What you are more likely to see in the halls of your classroom building