Response Sets

When participants answer questions, psychologists typically assume that they are responding to the content of the questionnaire items. For example, when participants are confronted with the question "I have never felt like smashing things," psychologists assume that participants think of all the times when they were angry or frustrated and then recall whether on those occasions they have ever felt like smashing or actually did smash something. Psychologists also assume that participants make a

Honest

The psychologist concludes he or she is . . .

Dishonest

The person being tested really is being . . .

Honest Dishonest

The person being tested really is being . . .

Honest Dishonest

Honest

Dishonest

Correct

False positive "incorrect"

False negative "incorrect"

Correct

Two ways to make a mistake when deciding whether a person was faking his or her responses to a personality questionnaire.

Two ways to make a mistake when deciding whether a person was faking his or her responses to a personality questionnaire.

deliberate and conscious effort to consider the content of the question and then answer "True" or "False" to honestly reflect their behavio . This assumption may sometimes be incorrect.

The concept of response sets refers to the tendency of some people to respond to the questions on a basis that is unrelated to the question content. Sometimes this is also referred to as noncontent responding. One example is the response set of acquiescence, or yea saying. This is the tendency to simply agree with the questionnaire items, regardless of the content of those items. Psychologists counteract acquiescence by intentionally reverse-scoring some of the questionnaire items, such as an extraversion item that states, "I frequently prefer to be alone." Extreme responding is another response set, which refers to the tendency to give endpoint responses, such as "strongly agree" or "strongly disagree" and to avoid the middle part of response scales, such as, "slightly agree" or "slightly disagree."

Many personality psychologists worry about the ef fects of response sets on the validity of questionnaire information. If a participant is responding not to the content of the questions but on another basis, then his or her answers do not reflect the aspec of personality being measured. Response sets may invalidate self-report measures of personality, so psychologists have looked for ways to detect and counteract the ef fects of noncontent responding.

The response set known as social desirability has received the greatest amount of research and evaluation by personality psychologists. Socially desirable responding is the tendency to answer items in such a way as to come across as socially attractive or likable. People responding in this manner want to make a good impression, to appear to be well adjusted, to be a good citizen. For example, imagine being asked to answer "T rue" or "False" to the statement "Most of the time I am happy ." A person might actually be happy only 45 percent of the time yet answer "T rue" because this is the well-adjusted thing to say in our culture. People like happy people, so the socially desirable response is "Y es, I am happy most of the time." This is an example of responding not to the content of the item but to the kind of impression a "T rue" or "False" answer would create, and it represents a response set.

Some rare individuals, like the late Mother Teresa of Calcutta, might score high on social desirability because they are in fact truly good, not because they want to create a good impression of themselves by lying on a personality questionnaire.

There are two views regarding the interpretation of social desirability . One view is that it represents distortion or error and should be eliminated or minimized. The other view is that social desirability is a valid part of other desirable personality traits, such as happiness, conscientiousness, or agreeableness. We will first consider ho psychologists have viewed social desirability as distortion.

Viewing social desirability as distortion does not assume that the person is consciously trying to create a positive impression. A social desirability response set may not actually be an outright ef fort to distort responses and, so, is dif ferent from outright faking or lying. Some people may simply have a distorted view of themselves or have a strong need to have others think well of them. For this reason, most psychologists have resisted calling this response set "lying" or "faking" (cf. Eysenck & Eysenck, 1972, for a dif ferent opinion). Nevertheless, many personality psychologists believe that socially desirable responding introduces inaccuracies into test scores and should be eliminated or controlled. If you wanted to know how happy a person perceives him- or herself to be, for example, you would want to have an accurate measure of his or her true level of happiness, not one that is contaminated by a need to create a good impression.

One approach to the problem of socially desirable responses is to assume that they are erroneous or deceptive, to measure this tendency , and to remove it statistically from the other questionnaire responses. There are several social desirability measures available to the personality psychologist. Several items from a popular measure developed by Crowne and Marlowe (1964) are presented in T able 4.1. Crowne and Marlowe thought of social desirability as reflecting a need for approval

Table 4.1 Crowne/Marlowe Scale for Measuring Social Desirability instructions: Listed below are a number of statements concerning personal attitudes and traits. Read each item and decide whether the statement is true or false as it pertains to you personally.

True False

1. I'm always willing to admit it when I make a mistake. ___

3. I never resent being asked to return a favor. ___

4. I have never been irked when people expressed ideas very different from my own. ___

5. I have never deliberately said something that hurt someone's feelings. ___

7. There have been occasions when I took advantage of someone. ___

8. I sometimes try to get even rather than forgive and forget. ___

9. At times I have really insisted on having things my own way. ___

10. There have been occasions when I felt like smashing things. ___

Source: from Crowne & Marlowe, 1964.

and they published the social desirability scale in their book The Approval Motive. Looking at the items on their scale, you can see that they typically refer to minor transgressions that most of us have committed, or inadequacies that many if not most of us suf fer from. In addition, some items refer to almost saintlike behavior . To the extent that a person denies common faults and problems and endorses a lot of perfect and well-adjusted behaviors, he or she will get a high score on social desirability. A person's score on social desirability can be used to statistically adjust his or her scores on other questionnaires, thereby controlling for this response set.

A second way to deal with the problem of social desirability is by developing questionnaires that are less susceptible to this type of responding. For example, in selecting questions to put on a questionnaire, the researcher may select only the items that have been found not to correlate with social desirability . This approach allows the test maker to build in a defense against the problem of social desirability during the process of constructing a questionnaire.

A third approach to minimizing the ef fects of socially desirable responding is to use a forced-choice questionnaire format. In this format, test takers are confronted with pairs of statements and are asked to indicate which statement in each pair is more true of them. Each statement in the pair is selected to be similar to the other in social desirability, forcing participants to choose between statements that are equiva-lently socially desirable (or undesirable). The following items (see bottom of page 115) from the Vando Reducer Augmenter Scale (Vando, 1974) illustrate the forced-choice format: Which would you most prefer (a or b)?

A Closer Look Integrity Testing

Polygraph exams formerly were widely used in employment screening until they were banned by Congress in 1988from use in private sector employment settings. The government, however, still uses polygraphs in employment screening as well as periodic honesty verification of persons in sensitive positions. In fact, the U.S. government runs several training institutes that certify persons to administer standard polygraph exams.

Polygraph exams formerly were widely used in employment screening until they were banned by Congress in 1988from use in private sector employment settings. The government, however, still uses polygraphs in employment screening as well as periodic honesty verification of persons in sensitive positions. In fact, the U.S. government runs several training institutes that certify persons to administer standard polygraph exams.

Throughout history, employers have been concerned about employee theft. Such thefts could be avoided or at least minimized if there were a way to tell whether a person was generally honest or dishonest before hiring him or her. Over two centuries ago, the Chinese developed a test to determine whether a person was lying. The test consisted of asking the suspect a question, waiting for the answer, and then placing rice powder in the suspect's mouth. If the suspect could not swallow the rice powder, it was viewed as a sure sign that he or she was lying. This may sound like superstition, but, if you think of the dry mouth that usually accompanies nervousness, then there might be some face validity to this early lie detection technique.

The modern lie detector, a polygraph, is a mechanical device that relies on psychophysiological measures, such as heart rate, respiration, and skin conductance (see Chapter 7). The use of physiological measures for lie detection started early in the 1900s in the United States. The idea behind this approach is that physiological measures may be useful in detecting the nervous arousal (e.g., guilt feelings) that often accompanies lying. The origin of the modern lie detection machine is shrouded in mystery. Some attribute it to a police officer from Berkeley, California, named Larson, who constructed the prototype of the multi-channeled polygraph between 1917 and 1921 and also published a manual on how to use the machine. Others trace the idea of using psychophysiological recordings—in particular, systolic blood pressure—to measure deception in laboratory and legal settings to William Moulton Marston, who worked on this problem while he was a graduate student at Harvard University from 1915 to 1921.

The lie detector gained widespread attention in the 1930s when it was introduced in the trial of Bruno Hauptman, who was accused of murdering the Lindbergh baby. Businesses began using the polygraph widely in the 1970s.

The polygraph was originally designed to detect guilt reactions arising from denying specific criminal acts. However, many employers began to use polygraph and other so-called lie detector tests to screen potential employees for general honesty. That is, the original purpose was to assess a state (guilt), whereas the polygraph was often pressed into usage to assess a trait (honesty). At any rate, participants were connected to these devices and asked various incriminating questions, such as whether they had ever taken anything that did not belong to them. If they showed any signs of nervousness or arousal (e.g., increased heart rate or shallower breathing) they might not have been hired. Employers also routinely used lie detector tests to question employees who were already on the job. Fast-food chains were among the largest users of polygraph tests in employment settings during this era (1970 to 1988). Managers hired polygraphers to connect employees to these devices, then ask questions such as whether they had taken any hamburgers or money in the past few months. If the polygraphs indicated any signs of nervousness, the employee might have been fired.

Through the 1970s and 1980s, more than 3 million polygraph tests were administered each year in the United States alone (Murphy, 1995). If you went into a large class of college students in the 1980s and asked if anyone had ever taken a polygraph exam, it was common to see at least a couple of hands go up for every hundred or so persons. Most said that they took the polygraph test as part of an employment screening procedure, often when applying for jobs in fast-food outlets.

A scientific evaluation of the polygraph as a lie detector was undertaken in 1983 by the U.S. federal government's Office of Technology Assessment. Its report concluded that there was no such device as a lie detector. Technically, this is true, as the polygraph detects physiological arousal, and sometimes lying is not accompanied by physiological arousal. In addition, sometimes physiological arousal is not accompanied by lying. The government evaluators also concluded that none of the methods used for lie detection were foolproof and that there were several effective ways to beat the device. Moreover, the polygraph's use in employment settings to screen for honesty trait may have resulted more in employment discrimination than in honesty detection.

In 1988, the U.S. Congress banned the use of the polygraph for most employment purposes in the private sector. Interestingly, the government still uses polygraphs for employee selection in several government service branches, such as the Secret Service, the CIA, the FBI, the DEA, Customs, and even the Postal Service. The government also maintains several polygraph schools, where people go to be trained in the use of the polygraph. In the private sector, however, the use of the polygraph in employment settings is highly restricted at this time.

This leaves the private sector employer with no mechanical means for detecting whether potential employees are honest or not. However, since the ban on polygraphs, many publishing companies have developed and promoted questionnaire measures to use in place of the polygraph (DeAngelis, 1991). These questionnaires, called integrity tests, are designed to assess whether a person is generally honest or dishonest. Many of these tests are considered to be reasonably reliable and valid and, so, may be legally used for employment screening (DeAngelis, 1991). Integrity tests measure attitudes related to one or more of the following psychological constructs: tolerating others who steal, beliefs that many others engage in theft, rationalizations that theft may be acceptable, interthief loyalty, antisocial beliefs and behaviors, and admission to stealing in the past. These tests typically consist of two parts. The first part measures attitudes toward theft, e.g., beliefs concerning the frequency and extent of theft, whether or not theft should be punished and how severely, and ruminations about theft. The second part concerns admissions regarding theft and other wrongdoing. Applicants are asked to describe the frequency and amount of theft and other illegal or counterproductive activity they engaged in on past jobs. Test items that make up integrity tests are clearly assessing job-related content (e.g., "Will everyone steal at work if the conditions are right?"; "Do you believe you are too honest to steal at work?"; "Do you think it is humanly possible for the average person to be completely honest on the job?" etc.).

A recent review of integrity questionnaires (Ones & Viswesvaran, 1998) looked at the use of these tests in organizations. They concluded that the measures are reliable (have test-retest correlations in the range of .85). There has been a great deal of validity research showing that integrity test scores can predict theft behavior. Questionnaire integrity tests have been found to predict the following theft criteria: (a) supervisors' ratings of employees' dishonesty, (b) applicants who are likely to get caught stealing once hired, (c) applicants who have a criminal history, and (d) applicants who are likely to admit theft in an anonymous testing situation. Longitudinal studies also demonstrate the impact of integrity tests. In one study, a group of convenience stores started using an integrity test to select employees and experienced a 50 percent reduction in inventory shrinkage due to theft over an 18-month period. A home improvement center chain also reported similar reductions in inventory loss after starting an integrity testing program.

When assessed against the big five traits—a widely accepted theory of personality known as the Big Five model (see Chapter 3, p. 82)—integrity appears to be a combination of high conscientiousness, high agreeableness, and low neuroticism. Moreover, these researchers found that integrity tests showed good predictive validity for absenteeism, counterproductive behavior on the job, violence at work, and theft on the job. They concluded that the concept of integrity has an important role to play in theories of job performance and counterproductivity in organizations, and that integrity tests can be valuable additions to typical measures used in employee selection (e.g., background checks, letters of reference).

2. a. eat soft food b. eat crunchy food

3. a. continuous anesthesia b. continuous hallucinations

4. a. a job that requires concentration b. a job that requires travel

If one answers all bs, this scale measures the preference for arousing or strong stimulation. The two choices presented in each item are of approximately the same value in terms of social desirability . Consequently, participants must decide on an answer based on something other than social desirability . They should respond to the content of the item and hence provide accurate information about their personalities. Other scales that use the forced-choice format to control for social desirability have been developed by Crandall (1991) and Buss et al. (1992).

Although many psychologists view socially desirable responding as error and as something to be avoided or eliminated, others see it as valid responding. Psychologists who subscribe to this point of view consider social desirability to be a trait in itself, one that is correlated with other positive traits, such as happiness, adjustment, and conscientiousness. These psychologists have argued that being mentally healthy may, in fact, entail possessing an overly positive view of oneself and one's abilities. In her book Positive Illusions, social psychologist Shelly Taylor (1989) summarizes a good deal of research suggesting that positive and self-enhancing illusions about the self, the world, and one's future can promote psychological adjustment and mental health. In a recent summary of this position, Taylor et al. (2000) review research that finds that unrealistic beliefs about the self (positive illusions are related to better physical health, such as slower progression of disease in men infected with HIV. If psychologists were to measure such positive illusions in the form of social desirability , and remove them from other personality measures, they might, in effect, be throwing the baby out with the bathwater . That is, social desirability may be part of being high on various trait measures of adjustment and positive mental health.

Work on social desirability has attempted to disentangle self-deceptive optimism from impression management. Psychologist Delroy Paulhus has developed a social desirability inventory, called the Balanced Inventory of Desirable Responding, which contains two separate subscales (Paulhus, 1984, 1990). The Self-deceptive Enhancement subscale was designed to tap self-deceptive overconfidence and contains item such as "My first impressions of other people are always right." The Impression Management subscale was designed to measure the tendency to present oneself favorably , as in the distortion interpretation of social desirability , and contains items such as "I don't gossip about other people's business." This subscale was intended to be sensitive to self-presentation motives, such as those that lead someone to want to create a good impression in others. In one study , the Impression Management subscale was strongly affected by instructions to the participants to fake good or bad, whereas the Self-deceptive Enhancement subscale, the part that measures overconfidence and pos itive illusions, was hardly af fected at all by these instructions (Paulhus, Bruce, & Trapnell, 1995). The Impression Management subscale might thus be sensitive to changes in self-presentation strategies, as might occur in job application settings or parole hearings (Paulhus, Fridhandler, & Hayes, 1997).

Anxiety and Depression 101

Anxiety and Depression 101

Everything you ever wanted to know about. We have been discussing depression and anxiety and how different information that is out on the market only seems to target one particular cure for these two common conditions that seem to walk hand in hand.

Get My Free Ebook


Post a comment