Misconceptions of psychology

Misconceptions of psychology Numerous studies have shown undergraduate students and everyday people to have a range of misconceptions about psychology. This study examined misconceptions among one group of first year psychology students, and another group of first year engineering students in order to determine whether psychology students perform significantly better than students of other academic disciplines in regards to their knowledge of psychology. A quasi-experimental independent-groups design was used, with the independent variables in this study having two levels, or whether the participant was either a psychology or engineering student and the dependent variable of the number of correct questions the individual achieved on the questionnaire. Results of this study showed that students from the two disciplines differ in their misconception and knowledge of psychology, with descriptive statistics and analyses of an independent groups t-test and a point biserial correlation showing a significant difference between the two groups. Conclusively, this study suggests that psychology students do have fewer misconceptions compared to engineering students, while further stating an alternative explanation and critically analysing procedures used to determine the produced results. Further research of in this area could focus on testing the misconceptions of a more varied sample, administer a different type of questionnaire, while also extending such research to a range of groups such as psychologists, doctors, teachers, or students of numerous disciplines. Misconceptions of psychology: A comparative study between psychology and engineering students Psychological research, regardless of the specific topic of study, is based upon the same scientific principles as the other hard sciences like physics and chemistry. Nevertheless, psychology has a long history of fighting a commonly held perception that it does not qualify as a scientific discipline and that it generates knowledge that is mere common sense. In many cases, however, common sense leads to an incorrect appreciation of phenomena that have been scientifically investigated by psychologists. Several studies have shown that undergraduate students and lay people alike hold many misconceptions about psychology. These misconceptions have been demonstrated in the United States and England and in several different courses of study. McCutcheon, Furnham, and Davis (1993) asked if there was a significant difference in the prevalence of misconceptions about psychology between American and English university students. They administered a 65 item multiple-choice questionnaire and found that English students generally did better than American students, though neither group answered more than half of the questions correctly. While the authors offered no clear explanation for either the poor performance or the difference between the two groups, they speculated that selective reporting in the mass media may contribute to both the formation of misconceptions and their resolution. Martin, Sadler, and Baluch (1997) tested undergraduate students from psychology, sociology, business, English and engineering from Middlesex University, U.K. on their general knowledge of psychology. Questions on their test invited common sense but incorrect answers. Out of a possible score of 38, psychology students scored highest with an average score of 17.08 while engineering students scored lowest with 11.57. Martin et al. also found that engineering students were least likely to regard psychology as one of the hard sciences or even one of the social sciences. The present study followed up previous reports of students generally poor performance on tests of misconceptions about psychology. Specifically, we repeated a portion of Martin et al.s (1997) study of misconceptions about psychology. We tested two Australian undergraduate classes, one introductory psychology class and one introductory engineering class, on a true/false test of common misconceptions about psychology adapted from Best (1982).This way of replicating a study is known as conceptual replication. While not all details of the Martin et al. study were replicated, this study allowed researchers to answer the overall question of misconceptions in psychology from a different angle. For this to be considered an exact replication of the Martin et al. study, misconception tests should have been given to individuals from a range of both first and last year psychology, sociology, business studies, engineering and English students, compared to just first year psychology and engineerin g students. Additionally, this study did not use a multiple choice test of misconception, but rather assessed misconception using a true or false questionnaire and provided no questionnaire of student attitudes towards psychology. The present study aimed to determine whether psychology students perform significantly better than students of other academic disciplines in regards to their knowledge of psychology. It was hypothesized that psychology students would have the lowest number of misconceptions about psychology, while engineering students would express the most misconceptions. Method Participants The 641 participants in this study consisted of 282 first year, undergraduate psychology students and 359 first year, undergraduate engineering students. Of the two groups in the study, the psychology group consisted of 75 males and 207 females, with ages ranging from 16 to 59 years, and a mean age of 20.3 years and a standard deviation of 5.5 years. The engineering group consisted of 264 males and 95 females, with ages ranging from 16 to 59, and a mean age of 20.0 years and a standard deviation of 5.3 years. The experiments were conducted at a university during Week 2 tutorials in the second half of the university year. Students were recruited by means of whether they were enrolled in either PSYC1040 or ENGG1000, two courses offered at the university. Design and Materials For the present study, a quasi-experimental independent-groups design was used. Similar to a true experiment, this study set up two levels of the I.V. (independent variable) and measured its effect on the D.V. (dependent variable). In this case, formation of the two groups was established by random assignment, but also by criteria established prior to completing the study. The independent variable in this study had two levels, or whether the participant was either a psychology or engineering student, while the dependent variable was the number of correct questions the individual achieved on the questionnaire. The key difference in a quasi-experiment is that participants in this study were tested on their knowledge and misconceptions according to characteristics they already acquired. The experiment used a 36-item true or false questionnaire (see Appendix A) to measure participants misconceptions and knowledge about psychology. The statements were drawn from a previous study conducted by Best (1982), which additionally suggested these statements appeared in a range of contemporary psychology textbooks and other similar questionnaires. Each statement consisted of a short sentence describing a simple scenario or event (e.g. psychiatry is a subdivision of psychology) and for which each participant was instructed to provide a true or false answer. Procedure All 641 participants received one copy of the questionnaire. Individuals were given the questionnaire during tutorials in Week 2 of Semester 2, 2010. Members of each tutorial group were assigned randomly by means of being there at the given time. Participants were instructed to complete each item individually and answer the statement as honestly as possible. The questionnaire was completed individually by the participants, without any interaction with fellow colleagues. Participants were given 10 minutes to fill in their answers, after which they were asked to swap their questionnaires with the person sitting next to them. Each participant then counted how many of the answers were correct and wrote down a score out of 36. For all the questions, the answer was false. Questionnaires and scores for both PSYC1040 (see Appendix B, figure 3) and ENGG1000 (see Appendix B, figure 4) were collected for further analysis and publication of the raw data. Results Data for both groups was explored graphically through frequency tables, frequency polygons, and box-and-whisker plots and were statistically analysed by means of an independent groups t-tests and a squared point-biserial correlation coefficient. Table 1. Mean questionnaire score and standard deviation of PSYC1040 and ENGG1000 Descriptive Statistic Psychology Engineering Mean 20.674 12.262 Standard Deviation 5.246 4.030 Psychology students scores were the higher of the two groups, with a mean of 20.7 and a standard deviation of 5.25 (see Table 1). Engineering students, on average scored lower than psychology students, with a mean of 12.3 and a standard deviation of 4.0 (see Table 1). As further indicated in Appendix C, psychology students performed better, with clear indication scores ranged between 16 and 26, while the engineering students highest scores ranged from 9 to 18 correct answers. Relative frequencies (see Appendix C, tables 4 5) were greater for scores between 15 and 28 for the PSYC1040 group, while scores between 8 and 18 for the engineering students proved to be more relatively frequent. Similarly, Figures 2 and 3 (Appendix D) illustrate PSYC1040 participants to have performed better than ENGG100 students, as psychology students achieved a higher score of 20 (indicated by the peak in the graph) most frequently and compared to engineering students who achieved a high score of only 12 m ost frequently. Additionally, from the distribution illustrated, the highest score from the PSYC1040 group was 36, while the lowest score was 6. Contrastingly, engineering students performed worse, achieving a high score of 30 and a low score of 0. Systematically, psychology students scores (See Appendix E and Appendix G, Figure 5), showed a mode and median of 20 and a range of 30. Descriptive statistics (Appendix E) and the box and whisker display (Appendix G, Figure 6) for the engineering group indicate a mode of 11 and a median of 12, while the range proved to be 30. Descriptive statistics (Appendix E) for the two groups show a significant difference between the two groups. Psychology students in fact performed 8.41 points greater than engineering students. However, in order to obtain a difference which is statistically important, certain assumptions were made to in order to perform further statistical analyses. Three assumptions were made of the independent groups t-test performed. Firstly, it was assumed that scores of PSYC1040 were independent of scores by ENGG1000 students, that data collected is representative of normally distributed populations and that variances of the two groups scores are alike. Test scores of PSYC1040 can be assumed to be independent of test scores of ENGG1000, as individual test scores from PSYC1040 could in no way be related to test scores achieved by ENGG1000 students, as the two groups were different. By means of illustration using the frequency polygons (Appendix D), the second assumption can be proven correct. Whi le the two distributions of the sample look more skewed and peaked than a normal ideal curve, we can make a decision that indeed the two samples of scores are from normally distributed populations as distributions of scores are not too skewed or peaked to violate such an assumption. Variances of the two populations differ slightly, as variance for the psychology group was 27.525 and variance for the engineering group was 16.243. There is no great difference to assume the two differ, so as a result we accept the given data and proceed. An independent groups t-test (see Appendix H) revealed that psychology students (M = 20.67, SD = 5.25) performed significantly better than engineering students (M = 12.26, SD = 4.03), with t (639) = 22.92, p Discussion As predicted, psychology students performed best in the misconceptions of psychology test than engineering students. The independent t-test and point biserial correlation both support the hypothesis proposed. The independent t-test showed significance between the two groups. The difference is very unlikely to occur if the samples were of no difference in the population and as a result the assumption is rejected. In other words, we reject the null hypothesis () and accept the alternative hypothesis, or that we are accepting a difference between the means that is no equal to zero. In this case the mean for psychology test scores is larger than the mean for engineering test scores, so it can be concluded that studying psychology leads to significantly higher tests scores when it comes to misconceptions of psychology. Additionally, the squared point-biserial correlation coefficient also supports the predicted hypothesis. Indicating the proportion of variance in a two sample study, the proportion ranges from 0 (variable has no effect) to 1.0 (variables control scores and therefore 100% of all scored can be predicted accurately). Variables representing between 10% and 30% of variance, indicated a relationship. Therefore, it can be concluded that being an engineering or psychology student accounts for a significant amount of the variance in the test scores. The variable therefore plays an important part in determining the test scores on the misconceptions survey. This means that the study was significantly more accurate at predicting a participants knowledge of psychology by knowing the condition they were in as opposed to just utilising the overall mean score for both conditions. Alternatively, composition of the two classes may have affected the data obtained. In both classes a gender difference exists. The psychology group consisted of more females than males, while the engineering group consisted of more males than females. Uneven numbers of males and females allows the data to be slightly bias. While experimental logic states that if one variable is manipulated, and if everything is kept the same, then all differences we observe are due to the manipulation of that variable. Quasi- experimental studies, which use gender as a variable, cannot be easily manipulated, meaning we cannot change gender throughout the experiment and we cannot be sure everything else stays the same. In order to control for the factor of gender bias, it is important to ensure the samples are representative of both males and females, in both the psychology and engineering groups. Possibly, other types of groups could be tested, all of which should be representative of both genders. McCutcheon (1991) critics the questionnaire used in this study. The Test of Common Beliefs has been constantly criticised on the basis that many items are ambiguously written. It has been argued that making the correct answers to all items false exposed individuals to accept without protest and gives many the opportunity to guess the answer by chance. Furthermore, McCutcheon suggests, items on several misconception tests are not widespread, meaning the test only focuses on a specific set of topics. Several studies such as Vaughan (1977), Lamal (1979), Gardner and Dalsing (1986) and Griggs and Ransdell (1987) all found that only a fraction of their questionnaire items were answered and only by half of their participants. It is possible to come to the conclusion that students which are beginners in psychology do not have many misconceptions about the subject. McCutcheon also proposes a possibility that true-false surveys are perhaps not the best at measuring misconceptions. In order to counter for this, McCutcheon steers away from using the true-false test of misconceptions, including several questions related to participant interest. Overall, results of this study suggest that in fact psychology students do have fewer misconceptions about psychology. It may be that other students of other disciplines are not aware of the knowledge required to overcome such misconceptions and therefore may even dismiss psychology as being a hard science. Further research in the area could include a study of misconceptions based on a larger, varied sample, while also using a questionnaire related to participant interests instead of a true or false questionnaire. Furthermore, future studies could also integrate participants who may be more experienced in the field, such as qualified psychologists in order to further analyse inter-disciplinary comparison of attitudes about psychology.