Using Daily Horoscopes to Demonstrate Expectancy Confirmation
Prior expectations can lead a person to process information in a biased manner such that the expectations are confirmed. Despite its prevalence in everyday judgments, people lack insight into expectancy confirmation processes. The classroom demonstration we describe uses daily horoscopes to demonstrate the powerful effect that expectations can have on judgments. More students reported that the events of a recent day most closely matched the prediction for their particular astrological sign when the astrological signs were presented along with the predictions than when they were not. Students reported that the demonstration was interesting, and 95% of the students correctly answered a question regarding the phenomenon on a subsequent examination.
Humans have difficulty perceiving the world in a truly objective manner because their expectations influence their judgments. A wealth of research has shown that information processing biases can lead to confirmation of expectancies (Darley & Fazio, 1980; Merton, 1948; Rosenthal & Jacobson, 1968). Expectancies influence the perceptual encoding of information (von Hippel, Sekaquaptewa, & Vargas, 1995), the retrieval of information (Snyder & Uranowitz, 1978), and one's behavior (Snyder, Tanke, & Berscheid, 1977). Finally, when one has a hypothesis, that person often searches for confirming rather than disconfirming information leading her or him to find evidence that supports the hypothesis (Snyder & Swann, 1978).
Psychology instructors, of course, understand how expectancy confirmation can influence judgments. They often present to students intriguing research examples detailing the biased processing of participants in some distant and past experiment. However, although it is easy to understand how expectations affected participants in these experiments, it is sometimes difficult to appreciate the ease with which everyday judgments can be affected. Indeed, the power of expectancy confirmation comes from this lack of insight. Thus, the instructor's task is to provide students with insight into how expectancies can influence everyday judgments.
A topic that has been used with considerable success in demonstrating a variety of scientific methods and psychological concepts is the scientific inquiry of astrology (Franknoi, 1988; Jones & Zusne, 1981; Kelly, 1989; Ward & Grasha, 1986). Although astrology has not been used to demonstrate expectancy confirmation, it might be particularly appropriate for that task. When daily horoscopes are compared to actual daily life events, it seems likely that a person's expectations that horoscopes are accurate might lead that person to find confirming evidence. The demonstration discussed in this research has two advantages that are directly related to the problems associated with expectancy confirmation. First, many students will have the opportunity to experience expectancy confirmation firsthand, and all students will directly observe its effects. Second, as a consequence of observing expectancy confirmation, students should gain an appreciation for the biasing effects of expectancy confirmation in scientific research and everyday judgments.
Method
Preparation
Two days before the class, the instructor collected daily horoscopes from a newspaper and prepared three overhead transparencies. The first, based on the horoscopes for the day that was 2 days before the class, included the zodiacal sign, the range of birth dates that correspond to that sign (for those unfamiliar with their zodiacal sign), and the predictions for that day's events. The second was identical to the first except that it was based on the horoscopes for the day that was 1 day before the class. The third, also based on the horoscopes for the day that was 1 day before the class, included only the predictions for that day's events in scrambled order labeled with the numbers 1 through 12 rather than zodiacal signs and birth dates. If daily horoscopes are not available, personality descriptors for each of the zodiacal signs could be substituted (although some students may know the traits associated with each sign).
Procedure
To introduce the demonstration, the instructor feigned enthusiasm for the use of horoscopes as a means of predicting life events. The instructor then posed the question as a scientific hypothesis that should be tested empirically. He explained to the students that one way to test for accuracy is to compare the horoscopes' daily predictions to the actual events experienced on that day. Students thought about what happened to them 2 days before. Then, the instructor presented the first overhead. The task of the students was to find the prediction (out of all 12 horoscopes) that best described what happened to them 2 days before. Once they found a prediction, each student passed forward a small slip of paper with their actual zodiacal sign and the sign of the best prediction of what happened to them 2 days before. The instructor then read each slip to the class to determine how many students thought that the best prediction also happened to be their actual zodiacal sign. This number was coded as the number of matches. Depending on the size of the class and the amount of time available, this step of the demonstration could also be completed by a show of hands. A percentage was computed by dividing the number of matches by the total number of students in the class and was then compared to the percentage one would expect to find by chance responding (8.3%).
After simulating excitement about the potential for astrology to predict life events at better than chance levels, the instructor suggested that one more test was necessary. He then presented the third transparency, which showed only the order-scrambled daily event predictions without the accompanying birth dates or zodiacal signs. This time, students found the one prediction that best matched the events of 1 day before the class. Once the students identified a prediction, each student passed forward a small slip of paper with their actual zodiacal sign and the number of the prediction that best fit what happened to them 1 day before. The instructor then presented the second transparency, which included the actual predictions from the previous day as well as the accompanying zodiacal sign. The instructor read each slip to the class to determine (by referring to the second transparency) how many students thought that the best prediction also happened to be their actual zodiacal sign. Again, students could be asked to perform this step on their own with the number being determined by a show of hands. After computing a percentage, it was compared to the percentage one would expect by chance responding.
Results
Demonstration Results
The instructor collected the data from four recent introductory psychology sections. When students indicated that the prediction that best fit the events of the day was also the prediction for their actual zodiacal sign, it was recorded as a match. When students indicated that the prediction that best fit the events of the day was not the prediction for their actual zodiacal sign, it was recorded as a mismatch. The numbers of matches and mismatches were then entered as the observed data in a chi-square where the expected chance data were 1 match for every 11 mismatches.
When the predictions were labeled with zodiacal signs and birth dates, students reported matches (34%) at a level significantly greater than the 8.3% expected by chance responding, X2( 1, N = 128) = 101.84, p < .001. When the predictions were labeled only with numbers and not zodiacal signs or birth dates, students reported matches (13%) at a level not significantly different than the 8.3% expected by chance responding, X2( 1, N = 128) = 3.58, p > .05.
Student Evaluations
To evaluate the demonstration, students from two of the introductory psychology sections responded to a short questionnaire following the demonstration and a discussion of how the demonstration was relevant to the topic of the day. The questionnaire included four rating scale items and two open-ended items. Students responded on a 5-point scale with 1 (strongly agree) and 5 (strongly disagree) as the endpoints and 3 (neutral) as the midpoint, to the following items:
- This exercise was an interesting learning experience.
- Doing this exercise was a valuable way to learn.
- This exercise helped me to understand how expectations can influence judgments.
- Would you recommend this exercise for future classes?
The two open-ended items were as follows:
- What specifically did you learn through the exercise?
- Please write any additional comments below.
The instructor asked the students not to write their names on the questionnaires and assured them that their responses would be anonymous.
For the four rating scale items, the means computed from the 73 students were all between 1 (strongly agree) and 2 (agree: interesting M = 1.82, SD = .56; valuable M = 1.67, SD = .58; understand M = 1.61, SD = .57; recommend M = 1.90, SD = .63). In fact, no student answered disagree or strongly disagree to any of the questions. In responding to the open-ended question assessing what students learned as a result of the demonstration and discussion, 67 of 73 students (92%) clearly expressed the idea that expectations can influence judgments. Additionally, on an exam given several weeks later, students listed the weaknesses of observational research (the demonstration was originally presented to the students in this context). In responding to the question, 69 of 73 students (95%) correctly answered that one weakness was that the expectations of the observer could influence what observations were reported. Of course, the demonstration is only one of many sources of learning (including textbook readings and reviews of content material) that may have contributed to the students' examination performance.
Discussion
The demonstration appears to be an effective technique for teaching students about expectancy confirmation. Students found it to be an interesting and valuable way to learn. Moreover, on two separate assessments after the activity, a vast majority of students demonstrated their knowledge of the expectancy confirmation phenomenon. Furthermore, the demonstration can be used in many different psychology courses to illustrate a wide variety of theoretical and applied examples of expectancy confirmation including the weaknesses of observational research, the double-blind design, the self-fulfilling prophecy, and the fallacy of personal validation of personality assessment (Forer, 1949; Hyman, 1989). Although the demonstration gives students firsthand experience and observation into expectancy confirmation, it is also important for students to interpret the findings after the demonstration is conducted. Specifically, students need to make the connection between the demonstration results and the tendency for people to search for and find expectancy-confirming information. After a firm understanding of expectancy confirmation becomes apparent, a number of potential routes for discussion could follow. First, the instructor could generate a discussion focusing on the different processes by which expectancies are confirmed (e.g., encoding bias, self-fulfilling prophecy, etc.) . In a more advanced class (e.g., social psychology), the class could debate whether these processes are driven by wants and desires or by pure information processing biases. Second, the instructor could ask students to generate other real-world situations in which expectancy-confirmation processes might apply. Third, in a research methods course, the instructor could engage students in a discussion regarding the advantages and disadvantages of within-subjects and between-subject designs given that the demonstration makes use of the former whereas an actual experiment would most likely make use of the latter. Finally, one could use the demonstration as a starting point for an investigation into pseudoscientific claims (e.g., the ability of palm readers and psychics to accurately assess personality) and into speculations as to how those claims can best be investigated scientifically (Hyman, 1989).
1. We thank Lucy Bohne, Jennifer Grieme, Terell Lasane, Richard Platt, and several anonymous reviewers for their helpful comments on this article.
2. Send correspondence to Geoffrey D. Munro, Department of Psychology, St. Mary's College of Maryland, 18952 East Fisher Road, St. Mary's City, MD 20686; e-mail: gdmunro@osprey. 8mcm.eau.
References
Darley, J. M., & Fazio, R. H. (1980). Expectancy confirmation processes arising in the social interaction sequence. American Psychologist, 35, 867-881.
Forer, B. R. (1949). The fallacy of personal validation: A classroom demonstration of gullibility. Journal of Abnormal and Social Psychology, 44, 118-123.
Franknoi, A. (1988). Horoscopes versus telescopes: A focus on astrology. Universe in the Classroom, 11, 1-4.
Hyman, R. (1989). The elusive quarry: A scientific appraisal of psychical research. Buffalo, NY: Prometheus Books.
Jones, W. H., & Zusne, L. (1981). Teaching anomalistic psychology. Teaching of Psychology, 8, 78-82.
Kelly, J. T. (1989). Astrology, science, and common sense. Science Activities, 26, 16-19.
Merton, R. K. (1948). The self-fulfilling prophecy. Antioch Review, 8, 193-210.
Rosenthal, R., & Jacobson, L. F. (1968). Pygmalion in the classroom. New York: Holt, Rinehart, & Winston.
Snyder, M., & Swann, W. B., Jr. (1978). Hypothesis-testing processes in social interaction. Journal of Personality and Social Psychology, 36, 1202-1212.
Snyder, M., Tanke, E. D., & Berscheid, E. (1977). Social perception and interpersonal behavior: On the self-fulfilling nature of social stereotypes. Journal of Personality and Social Psychology, 35, 656-666.
Snyder, M., & Uranowitz, S. W. (1978). Reconstructing the past: Some cognitive consequences of person perception. Journal of Personality and Social Psychology, 36, 941-950.
von Hippel, W., Sekaquaptewa, D., & Vargas, P. (1995). On the role of encoding processes in stereotype maintenance. Advances in Experimental Social Psychology, 27, 177-254.
Ward, R. A., & Grasha, A. F. (1986). Using astrology to teach research methods to introductory psychology students. Teaching of Psychology, 13, 143-145.
~~~~~~~~
By Geoffrey D. Munro, St. Mary''s College of Maryland and James E. Munro, Edinboro University of Pennsylvania
Send correspondence to Geoffrey D. Munro, Department of Psychology, St. Mary's College of Maryland, 18952 East Fisher Road, St. Mary's City, MD 20686; e-mail: gdmunro@osprey. 8mcm.eau.
Copyright of Teaching of Psychology is the property of Sage Publications Inc. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
icons on the left to see existing comments.
0 General Document comments
0 Sentence and Paragraph comments
0 Image and Video comments
It’s really interesting to me that people look at something as simple as a horoscope and think that it can be accurate for their expectations for the day. Personally, when I look at a horoscope, I find it sort of silly and generalized. It’s bizarre to me that people believe in astrology as a thing
New Conversation
Hide Full Comment
Looking at a horoscope, it’s just one big guess. As I said before, they seem very generalized. How can you know what your “daily horoscope” is based on the arrangement of the planets in our solar system. It’s just strange. I think this shows how random it is.
New Conversation
Hide Full Comment
If you’re thinking about how your day is going to go according to a certain specific thing, your brain will be influenced by that thing. When reading a horoscope, you can easily say, “oh that’s how my day is going to go” and be influenced in that way.
New Conversation
Hide Full Comment
General Document Comments 0
