Measuring Consistency of Trait Attributions

If you were asked to describe your best friend, you would probably do so by listing a set of adjectives-such as loyal, generous, funny, easygoing, and so on-that you believe represent this person's personality. In fact, we all make these kinds of judgments automatically, even when we are first introduced to a person. Perhaps you have wondered how consistent people are in their judgments of others. When people meet you, are they usually left with the same first impression? The following experiment will give you some idea about how consistent different people's judgments of another person really are.

Experimental Procedure

To do this experiment, you will need to go through some magazines looking for pictures of three people who, on first impression, strike you as having widely varying personality traits. For example, you might choose a picture of the "Marlboro man" (a handsome, strong-looking man), a factory worker, and a beautiful woman (perhaps in an advertisement for liquor or perfume). (Avoid using pictures of well-known politicians or entertainers, because your subjects are likely to know something about such public figures and to have already made judgments about their personalities.) When you have found three pictures, cut them out and number them 1, 2, and 3. You will use these three pictures as stimulus material in this experiment.

You will also need to make five photocopies of the Rating Form provided here and to enlist five cooperative subjects.

STEP 1 Give each subject a copy of the rating form and the three pictures you have selected and ask them to circle the word in each word pair that they think accurately describes this person's true personality. You will need to explain that, of course, they have no way of knowing what these people are really like, but they should make their best guess based on what they see in the pictures. Encourage the subjects to choose one word from every pair.

STEP 2 When the subjects have rated all three pictures, thank them for their help in providing the data for this experiment and collect their rating forms.

Analyzing the Data

STEP 1 Complete the Scoring Table by counting the number of subjects who chose each response. For example, if four of your subjects rated the person in Picture 1 as "intelligent" and one subject rated the person as "average," you would write "4" after "intelligent" and "1" after "average." Continue to fill out the table for each of the word pairs and for each of the three pictures.

STEP 2 You can now graph a personality profile for each of the people in your three stimulus pictures. Using the graph provided in Figure 13.1 and starting with the data for Picture 1, plot the number of subjects who used the first word in each pair to describe the person in the picture. Thus, you will plot the number of subjects who responded "intelligent" by putting a dot on the appropriate spot on the graph. (Do not plot the number who responded "average.") Next plot the number of subjects who responded "restrained." (Do not plot the number who responded "talkative.") Continue to plot the responses for the first adjective in each of the remaining eight pairs. Now connect the ten dots with a solid line. This is the personality profile for the person pictured in Picture 1.

STEP 3 Plot the data for Picture 2 just as you did for Picture 1. Connect these dots with a dashed line.

STEP 4 Plot the data for Picture 3, connecting the dots with a dotted line.

Analyzing the Results

After examining these three personality profiles, would you conclude that there is some consistency in the way your subjects viewed the pictures (did you plot lots of 0s, 1s, 4s, and 5s)? Or were your subjects split in their judgments of the pictures (mostly 2s and 3s)? Do the answers to these questions depend on which pair of adjectives you are considering? (That is, do some of the adjective pairs yield high or low numbers, whereas others produce more intermediate results?) Which adjectives seem to elicit the most agreement among your subjects?

Did your subjects' responses vary from one picture to another? (This would be seen as personality profiles that were quite different from one another.) Which adjective pairs produced the greatest discrepancies among the picture ratings? Would you have predicted these results?

Of course, the results you will get in this experiment are dependent on both the particular pictures you choose and the subjects you use. There is no way to anticipate which adjective pairs will prove to be consistent and which will be less consistent in their application to the different pictures.

Most people performing this experiment will, however, find that at least some of the pictures evoke some consistent adjectives. For example, if you used a picture of the "Marlboro man," you might expect that your subjects will frequently assign him the characteristics of "restrained, aggressive, and stable." In contrast, a picture of a housewife posing with a box of laundry detergent is often described as "average, talkative, pleasant, and easygoing." Do the pictures you've chosen evoke this kind of consistent response?

What conclusions can you draw from this experiment about the tendency for people to make consistent personality judgments on the basis of first impressions? Were your subjects' ratings generally in agreement? Do you think the first impression people have of you is fairly consistent? Finally, how would Walter Mischel's ideas about the existence of personality traits (discussed in the text on page 402) explain the results of this experiment?