Week 11 Comments for TA
10 Saturday Dec 2011
Posted Uncategorized
in10 Saturday Dec 2011
Posted Uncategorized
in09 Friday Dec 2011
Posted Uncategorized
inDr Fay Short recently briefly mentioned a study in her social psychology lectures that I remember being taught for A-level psychology & it was study that both shocked & intrigued me. So I have decided for my final blog to analyse the reliability and validity of this experiment because it does appear to be lacking in both.
The study “Is homophobia associated with homosexual arousal?” was conducted by Adams, Wright & Lohr (1996). The study involved 64, Caucasian, heterosexual males in the U.S, who had completed three tests. Only men who reported just heterosexual arousal and experiences were selected and then categorised depending on their score on the Index of Homophobia into either non-homophobic or homophobic men. These men were shown 3 videos showing; homosexual male sexual activity, homosexual female sexual activity and heterosexual sexual activity. Between each video participants’ arousal was measured by a ring that was placed over the penis and their self-reports of sexual arousal and degree of penile erection. The results indicated that the homophobic men showed a significant increase in arousal, according to the physiological measure, when watching the homosexual male video. Yet the participants’ subjective ratings showed participants’ rated themselves as being significantly more aroused by the heterosexual and female homosexual videos. According to psychoanalysts homophobia is a form of repressed latent homosexuality. (This is only a brief overview, read the report if your interested).
The basic criticism of this study, first of all, is how can a small sample of 64 men accurately represent the entire Caucasian male population in the U.S? Therefore the generalizability of the experiment is questionable and this needs to be kept in mind.
Whilst it has been found that the test-retest reliability of this measure of arousal is acceptable (O’Donahue & Letourneau, 1992), what actually is meant by acceptable and is this really good enough? Firstly, test-retest reliability is the correlation between the scores gathered from one sample and the same participants scores on the same test/task at a later time. In the case of this study it refers to the physiological measure of arousal, known as penile plethysmography, apparently when a participants’ arousal is measured and then at a later time under the same conditions measured again the scores should produce an acceptable correlation. To me the term acceptable implies that it is neither poor nor good, it’s just ok. According to Cohen a correlation of .50 is a large effect as it accounts for 25% of the total variance. However, what about the remaining 75%? Is it really acceptable for the remaining 75% to be down to error, something that is unaccounted for by the test? The participants’ anxiety could play a large part in their arousal levels and this is what I will go on to discuss next.
The biggest issue with this study, for me, is its validity. At first glance the face validity appears high, after all a physiological measure of arousal such as placing a ring (see the study for the technical terms) around the penis seems like an accurate way to measure arousal because it measures the electrical change in resistance. However, think about the last time that you took part in a SONA study did you feel anxious because I know that I did slightly? Now if I’m told that I’m going to be shown the same 3 videos as in this study & my sexual arousal is going to be measured, then my anxiety has drastically heightened. It is well known that an increase in anxiety can cause an increase in arousal, even at inappropriate times. Plus the participants’ were all heterosexuals and were informed that they would be watching a video of homosexual male sexual activity, it seems rather common sense that the homophobic men, more than the non-homophobic, became very anxious by this and were consciously trying to not become aroused, which does not help the situation. Therefore how do the authors of this study truly know that they were measuring arousal levels depicting latent homosexuality? After all the idea of latent homosexuality is that it involves repressed thought & actions in the unconscious mind, thus not possible to measure.
On the surface this study appears to be a scientific experiment with high reliability and validity, after all it uses physiological measurements which are often accepted without question. However, whilst there is no uncertainty that the ring placed around the penis does measure arousal it is unclear as to whether this is because of the homosexual male sexual activity or the fact that participants’ have been placed in an extremely uncomfortable, anxious situation, especially if they are homophobic.