Psychology theories of Attraction
Psychology Coursework: Introduction
Attraction, or physical attraction, can be described as what romantically draws people together.
The "Exchange Theory" developed by Thibut and Kelly (1959), explains that a relationship will be successful, dependant upon gains and loses by each person. This means it is rather like a "you scratch my back and I'll scratch yours" theory for a relationship.
The Matching Hypothesis is the tendency for people to select their partners who are alike them in looks, or education or social upbringing.
Relative to this theory, Towhey (1979) asked males and females how much they thought they would like a person whose photograph they had studied. They were also given biological information, such as height, age etc. The results showed that those who scored highest on the Macho Scale (dealing with sexist attitudes and stereotypes) were more influenced by physical attractiveness, compared to those scoring lower being almost unaffected. This doesn't necessarily support the Matching Hypothesis, but it shows that participants preferred more attractive partners.
Similarly, Dunbar and Waynforth (1995) conducted a study, analysing 900 "personal" adverts in 4 USA newspapers. They noted down details of all adverts, and then grouped the entire data dependant on gender. They found 44% of males wanted a physically attractive partner, opposed to 22% of women. These results can help with expectations on the study that will be performed. Perhaps the results will include the males rated higher in attractiveness that the women, based on Dunbar and Waynforth's results.
Silverman (1971) wanted to investigate the matching hypothesis theory in a more ecologically valid way. He assembled an observational team comprising of 2 males and 2 females as dating couples in natural settings, such as pubs and clubs. Each observer rated the attractiveness of the couples out of 5. it was found that there was a high degree of similarity in the couples' level of attractiveness. It was also noted that there was more physical intimacy between the couples that were most similar to each other. 60% of the most similar couples showed signs of intimacy such as holding hands etc. whilst only 22% of the couples labelled as dissimilar showed such behaviours. This therefore supports the matching hypothesis.
Walster (1966) conducted the "computer dance" study, to investigate the matching hypothesis. He took 752 student and they each filled out a questionnaire. They were told that they would be allocated a partner on the basis of their questionnaire, when in fact the pairings were made at random. Four judges rated their physical attractiveness, unknown to the students. The students were then paired up one evening at a dance. Walster found that the more attractive students were liked more than their partners than the less attractive ones.
This study doesn't support the Matching Hypothesis, as physical attractiveness was the most important factor in liking, above intelligence and personality. The study has low ecological validity, as in real life people are not allocated pre-assigned dates. However in some religions arranged marriages are common, however not in our culture. When the students were interviewed later on, they said they were more likely to date people of a similar physical attractiveness, which supports the Matching Hypothesis. The study has got some ethical problems, as the students didn't know they were being judged on appearance. This could have caused minor psychological harm in some students, who maybe very self-conscious. To overcome this, Walster could have got their permission to judge them in a consent form.
A study that supports the Matching Hypothesis even more is Murstein (1972). He took photographs of "steady or engaged" couples, and participants compared them to couples randomly paired up for the experiment.
Murstein found that couples that scored highly in similarity in attractiveness were the "real" couples. In this study, it should have been essential that consent form were handed to all participants, especially those that posed in photographs. This is because some participants could feel uneasy about being physically rated, and should have the right to withdraw at any time.
Murstein summarised his findings as:
"Individuals with equal market value for physical attractiveness are more likely to associate in an intimate relationship such as engagement than individuals with disparate values."
This study will be based upon Murtsein's experiment, investigating if the Matching Hypothesis is really true. As it can be seen in the previous studies above, not all support the Matching Hypothesis. Therefore it is proposed to test this theory. It is suspected that the Matching hypothesis is true, as Murstein conducted a very strong study to support it; he found true couples were equally as attractive as each other. As Murstein has given the study a direction for a hypothesis, it is one-tailed. Walster has also helped form a hypothesis, as in a real-life setting, with high ...
This is a preview of the whole essay
This study will be based upon Murtsein's experiment, investigating if the Matching Hypothesis is really true. As it can be seen in the previous studies above, not all support the Matching Hypothesis. Therefore it is proposed to test this theory. It is suspected that the Matching hypothesis is true, as Murstein conducted a very strong study to support it; he found true couples were equally as attractive as each other. As Murstein has given the study a direction for a hypothesis, it is one-tailed. Walster has also helped form a hypothesis, as in a real-life setting, with high ecological validity; his participants said that they would date others of the same attractiveness rating.
Aims
The aim of the study is therefore to investigate whether photos of individuals who are in couples will be ranked similarly. If they are then this would provide further support for the Matching Hypothesis, which states long-term couples will be of similar attractiveness.
Alternative Hypothesis
The alternative hypothesis states that there will be a positive correlation between the ranking of attractiveness for males and females in each couple as judged from photographs.
Null Hypothesis
The null hypothesis states that there will be no significant positive correlation found within the results, and any correlation found will be due to chance alone.
This is a one-tailed hypothesis, as it states the direction that a 5% level of significance will be applied to the results.
Method
Design
A partial representation of Murstein's (1972) study was undertaken to investigate the Matching Hypothesis. Attractiveness of individuals in couples were ranked using photographs. This is a non-experimental study, as it has no IV or DV, and the study is concentrated on the directional strength of two variables, (the ranking of attractiveness between photos of males and females in each couple). Therefore the study is a correlational study, where positive correlation will support the hypothesis and also the Matching Hypothesis.
Participants
In total, 36 participants were used. 16 participants were photographed to form 8 couples to be ranked. The remaining 20 took part in providing data. This consisted of 10 males and 10 females. They were all 17-18 years old, and the population sample were all students in college. An opportunity sampling method was used. This was the easiest way to get participants, as the researchers asked friends in the college to participate, knowing that there were a plentiful number of people to choose from around the time of the study. The participants were tested in an empty room, which was private and quiet, so no distraction could take place. The 8 adult couples who were photographed were volunteers, being friends or family of the researchers involved. The volunteers were all married, and all around the same age.
Materials
6 photos were used, 8 female and 8 male. To take these photos (displayed appendix 1), each person was given a consent form to sign (appendix 2).
A score sheet and consent form for the participants who ranked the photos were also used (appendix 3). On one side is the score sheet, and on the other is the consent form, this was doubled up as it saved time and space.
Procedure
A pilot study conducted in class formed the basis of this study; the researchers found that the quality of the photograph was important. So, the photos were taken against a plain background, from waist upwards. All were told to look pleasant and smile slightly, this was done so all photos could be judged the same and no-one would look especially more attractive due to the fact that they were smiling, or something similar. The photos were taken in daylight, so they could be seen clearly.
The study consisted of a table with 16 photos laid out on it. The male photos were labelled "A" to "H", as were the female photos. The photos were grouped in the male/female format. After signing the consent form, participants read the instructions on their score sheet (appendix 3) and any questions could be asked if necessary to the researcher standing nearby. The participants were told to rank, in order of attractiveness, the 8 different photos for the males, and then again for the 8 females.
This procedure used repeated measures, as it was repeated 20 times, using 20 scoring participants.
After the photos had been ranked, participants were debriefed by telling them what part they had played in the study, and also given the opportunity to withdraw their results again. The results were collated in a table to show ranking of attractiveness for each male and female couple (appendix 4).
Results
The raw data collected for the male and female photos is displayed in appendix 4. There is also a summary of the data in the tables below.
Table 1: the sum of the total rankings of attractiveness for the male and female photos in each couple.
Couple
Sum of attractiveness ranking
MALE
Sum of attractiveness ranking FEMALE
2
3
4
5
6
7
8
49
41
97
43
90
42
77
81
51
99
32
17
16
31
42
32
Looking at table 1, the couples' scores are quite similar. Comparing couple 1 and couple 7, the scores between the males and females are close, for instance 49 and 51, showing a similar level of attractiveness. The data is much clearer if the medians are taken from each couple and put into a table to summarise the data, making it easier to see patterns, trends and relationships.
Table 2: the median rank of attractiveness for male and female photos in each couple
Couple
Median rank MALE
Median rank FEMALE
2
3
4
5
6
7
8
2
7
5
7.5
5
4
4
3
5
7
6
6
6.5
2
Table 2 shows the median value of attractiveness for each pair. Again, the medians between the male and female in a couple are quite close, for instance in couple 1 the medians are 2 and 3. Couple 5 is equally as close, with 5 and 6 as their medians. There is only one very different result, in couple 6 where the medians are 1 and 6.5. this means that the Matching Hypothesis is proving to be true, as a close ranking score would mean similar levels of attractiveness. Below is a scatter graph to illustrate the medians in each couple. The line of best fit shows a positive correlation, which positively provides good evidence that the Matching Hypothesis is correct.
Treatment of results
The raw data was then analysed statistically, using a statistics test called Spearman's Rank Order Correlation Coefficient test.. This test signifies whether a positive correlation is down to chance. For instance, as the male rank of attractiveness increases, so does the female rank of attractiveness.
The Spearman's Rank Test was used because it can test the relationship between two variables, in this case the ranking of attractiveness for males, and ranking of attractiveness for females. This is suitable for the correlational study, as the aim was to find such a relationship. The Spearman's Rank test was also used, as it is suitable for ordinal data, as a ranking scaled was used.
Spearman's Rank is used in conjunction with the one tailed hypothesis to produce an answer to if the null hypothesis is accepted or not. The hypothesis formally mentioned is one-tailed as it predicts the direction i.e. a positive correlation. The calculation of Spearman's Rank can be found in appendix 5.
The calculated value of rs =0.24. In a 1-tailed hypothesis, at a probability of 5% the critical value (in table-appendix 6) is 0.643 at N=8 (where N is the number of couples used for the photos).
Since rs is less than the critical value, the Ho must be accepted. This means there is only a weak positive correlation. However, it can be said that it is a positive value (rs=0.24) and this supports the Matching Hypothesis in direction. However, to be accepted as a true results rather than chance, it must match or exceed the critical value. Also, a set significance of 5% was used to calculate the critical value; this means that there is a more than 1 in 20 chance that the Ho is correct. Therefore, although a positive correlation was found, it is not significant, and so the null hypothesis is accepted and the alternative hypothesis rejected.
Discussion
The null hypothesis was accepted. This stated that there would be no significant positive correlation found within the results, and any correlation found will be due to chance alone. The Spearman's Rank test resulted in a correlation rs=0.24 which is insignificant at a 5% level, in a one-tailed hypothesis.
However, looking at the results table, over half the couples used had an attractiveness ranking similar between the male and female. Only one couple, couple 6, had an anomalous result of 1 and 6. Therefore in relation to the majority of findings the Matching Hypothesis seems valid. However, the anomalous result prevented a significant result.
The anomalous result was down to couple 6, as their results (1 and 6) were so much further apart compared to all the other couples that it must be them alone who skewed the Spearman's Rank value. However, as the Spearman's Rank value was positive, and a positive correlation was established, it is decide that there must be some level of significance to the study, however the sample was too small for an anomalous result not to make a big difference.
Murstein's study on testing "real" and "pretend" couples found more significant results. Therefore if this study compared real and fake couples then a more significant result could be achieved, as the medians are quite close anyway. Due to this fact, perhaps the Matching Hypothesis has some truth. However it must be remembered that a series of factors come into the equation when choosing a life-long partner, and in this study only one factor was used.
The Exchange theory put forward by Thibut and Kelly (1959) can maybe be reinforced here, instead of the Matching Hypothesis. It states tat a relationship will be dependent upon gains and loses, rather than the attractiveness of each partner. As all our photographed volunteers were middle-aged and married, maybe it is that of the exchange theory that was made their relationships successful, as opposed to attraction. It is a well known fact that one "looses" their attractiveness as time wears on, so what once might have been a relationship brought together by attractiveness (which would be relevant to the Matching Hypothesis) is now a relationship bases on rewards and love, as in the Exchange Theory. So perhaps in future, the studies should focus on younger couples who are in their prime of attractiveness.
The main limitation to the study was the quality of stimulus material chosen. This is concerning the photographs used, and the differences between them. For example, a difference in light, background and the direction the participant is facing is to name a few. This could all be distracting to a person ranking the photos. The only way to overcome this is the researching firmly deciding on the specification on the photographs before being taken. Perhaps this may include preparing a specific place with specific light for all the photos to be taken.
The participants for the photos were chosen outside school so there would be less chance of any bias. For example, not many people would recognise the participants in the photos, so they would be under no obligation to rank them top. However, a small group of participants chosen still recognised some on the photographs, and this could have affected the true results. The only way to overcome this would be to choose either complete strangers for the photos, like in a shopping centre, or use the ranking participants who were sure not to recognise the photos. This would prove difficult to do, because opportunity sampling was used, and in the 6th form where the participants were chosen, the researchers did not know if people ranking would know those in the photos or not.
There was no age bias in the study, as all the people photographed were near enough the same ages, and those ranking the photographs were definitely the same age as they were all in the same year at school. However there might have been some age bias across generations, as young, 16-17 year old pupils are not likely to find 40-50 year old attractive, so by asking them to rank them could prove difficult.
Participants found it quite hard to rank photographs, and as a result participants would take quite a long time completing the study. To avoid this in future, perhaps a grading scale could be used as opposed to a rank order. Although this would be far easier for participants, when analysing the results it may be seen that participants have only used middle values such as 5 or 6, as opposed to the significantly higher or lower values such as 2 or 10.
There were not many ethical issues in the study, as most were taken care of by researchers giving the right to withdraw any results. However what may be seen as affecting the results were the male participants ranking the male photos. Many men feel uneasy about judging other men's looks. The only way to overcome this was to promise confidentiality and participants were reminded they could withdraw at any time
The anomalous result and the results of the couples whose medians were not so close may be explained by looking back on previous research. For example, if one person in a couple is drastically more or less attractive, then the relationship may not be just about looks. In Walster's study it was seen people attracted to each other on intellect, personality and interests, aside from physical attractiveness. This could therefore explain why an anomalous result was found.
It is true that attractiveness is very important when securing a partner, and this can be seen by walking into any beauty salon and seeing women and men getting beauty treatments, or having their hair styled. It is also true that there is a lot of pressure on people to look physically attractive, as seen in advertisements on clothes, make up, hair products and age-defying cream in magazines or on the television. Therefore the Matching Hypothesis does have logical truth to it. But people base relationships on more that looks, as shown in the findings.
Future studies could include gender, age and cultural differences. Culture would be very important, as each culture has a different criterion for physical beauty. For instance, the western world has an ever-changing opinion of what is considered attractive. A hundred years ago for women, plump, pale and curly hair was considered attractive, as it showed signs of wealth. However today a tan, straight blonde hair, long nails and a slim figure is what is seen as desirable. Compared to the ancient Orient where small feet were seen as attractive, there is a definite cause for research in cultural differences. Perhaps research could also be carried out into facial expressions, for instance smiling. People normally think of smiling as more attractive, so a study could be interesting. The hypothesis could be that if a person smiled they would be more attractive than if they didn't smile.
Also, rather than correlational studies, questionnaires could be used as well as interviews for more personal and in depth findings. This can really be used to understand each individual's opinion on what is attractive in a person.
Abstract
The main aims of the study were to test the Matching Hypothesis, which is the tendency for people to select partners who are alike them in looks, education and social upbringing, and see if it could be proven to be supported. A partial representation of Murstein's (1972) study that used photographs of couples which participants rated for attractiveness was used as a basis for the study.
6 photos were taken of individuals in 8 married couples, then an opportunity sample of 20 school pupils separately ranked the photos in accordance of physical attractiveness. The males were ranked separately from the females. If the Matching Hypothesis was true then the rankings within a couple should be very similar. The results were tabulated and then statistically tested for significance, using the Spearman's Rank test, and rs=0.24. There was an anomalous result that might have skewed the results. The correlation of 0.24 was not significant at the 5% level for a one-tailed hypothesis.
The conclusion of the study was that the Matching Hypothesis has some truth to it, as a weak positive correlation was obtained, however the results were statistically insignificant. However it was also decided that normally the Matching Hypothesis also considers intellect, social upbringing, personality as well as attractiveness, whereas the study conducted only considers one factor, so maybe other factors apart from attractiveness were relevant for the couples studied, or maybe the photos were not of sufficient high quality.
References
Bersscheid, E. & Walster, E. Physical Attractiveness. In L. Berkowitz (ed.,) Advances in Experimental Social Psychology. Vol.7, 1974. Excerpts.
Cardwell, M., Clark, L. & Meldrum, C. (1996). Psychology for A Level. HarperCollins Publishers Ltd.
Dunbar, R. (1995) Are you lonesome tonight? "New Scientist", 11 February, pp.26-31.
Huston, T.L. (1973) "Ambiguity of acceptance, social desirability and dating choice" Journal of Experimental Psychology, 9 pp.32-42
Murstein, B.I. (1972) "Physical attractiveness and marital choice" Journal of Personality and Social Psychology, 22, pp.8-12
Thibaut, J.W.; & Kelley, H.H. The social psychology of groups. New York: Wiley, 1959
Towhey, J.C. (1979) "Sex-role stereotyping and individual differences in liking for the physically attractive" Social Psychology Quarterly, 42, pp.285-289
Silverman, I. (1971) Physical attractiveness and courtship. Sexual Behaviour, September, 22-25.
Walster, E. and Walster, G.W (1969) "The matching hypothesis", Journal of personality and social psychology, 6, pp.248-53.
.
