BECKER ET AL. (2002)
THE CONTEMPORARY STUDY: TELEVISION & EATING DISORDERS IN FIJI

This study was carried out by an anthropologist, Dr Anne Becker. Anthropology is the study of other cultures so it has many similarities with psychology. Becker wanted to investigate the link between TV and the growth of eating disorders. Becker had been studying the culture of Fiji, a group of islands out in the Pacific, east of Australia, since the 1980s. When a Fijian TV station started broadcasting in the 1990s, this was an opportunity to see how TV affected the islanders.

This study is significant for students in other ways:
  • It shows how scientific research proceeds, because this is a study from the discipline of Social Anthropology that builds on ideas from the Learning Approach in Psychology.
  • It illustrates the usefulness of the natural (or quasi-) experimental method, taking advantage of the opportunity to study a naturally-occurring variable (the arrival of TV on a remote island)
  • It links in with your Learning Key Question, whether slim models on TV encourage anorexia
  • However, it raises important methodological issues about drawing conclusions from natural experiments

TELEVISION COMES TO FIJI

Fiji has a distinctive culture that is very different from the Western culture that was brought to it by European explorers. An unintended side-effect of the arrival of European settlers was the introduction of measles, a disease which caused many deaths among the Fijians who had little immunity.

In 1995, a TV station began broadcasting in Fiji, showing a mixture of British and American shows like E.R., Friends and Xena: Warrior Princess.
Some of the stars of '90s TV in Fiji: E.R., Friends, Xena and Melrose Place - and yes, that's Alex Kingston, Dr Who's River Song!
Up until this point, the Fijians had a preference for larger body types. Ranadi Johnston, a 'Miss Fiji' beauty queen winner, said slim women were traditionally seen as weak, explaining:

"People are always telling me to put on weight."
Anne Becker, who notes that eating disorders like anorexia had been almost unknown on Fiji, wondered if the arrival of Western TV, with its thin actors and models, would bring about a change in the way young Fijian women saw themselves and their bodies.

  • Notice that Dr Becker is taking advantage of something that is happening anyway – the introduction of Fijian TV is happening anyway and the researchers are not causing it or manipulating it. This is why the study is a natural experiment.
  • Also consider the implications of Dr Becker’s ideas. If the introduction of Western TV triggers an increase in anorexia, this would be similar to the introduction of measles in the 19th century.
Dr Becker explains her study in a short interview

BECKER'S STUDY
APRC

Aim

To find out if the introduction of Western TV into a community that has not previously experienced it would lead to an increase in unhealthy eating behaviour; specifically, to see if more young women would report purging (induced vomiting), dieting and body-dissatisfaction after the introduction of TV compared to earlier

IV

This is an Independent Groups design, since it studies two different groups of Fijian girls

One condition of the IV is a sample of girls studied in 1995, within a few weeks of TV being introduced; the other is a group of girls studied three years later in 1998. The girls were drawn from the same two schools and all were native Fijian islanders.

DV

Scores from the EAT-26 questionnaire about eating behaviour, with a score over 20 indicating signs of eating disorders (strict dieting, vomiting).

The researchers also gathered qualitative data on the girls’ self-image and attitudes towards purging, body-image, beauty and diet. This took the form of a semi-structured interview.

Sample

Two groups of Fijian schoolgirls in Years 5-7 (equivalent to Y11-13 in the UK), all around age 17 on average. There were 63 participants in 1995 and 65 in 1998.

Procedure

The schoolgirls completed the EAT-26 questionnaire.
The girls also received a semi-structured interview. They were all fluent in English but a translator was on hand to explain any cultural misunderstandings to the girls.

Both groups also answered questions on television in their home and measures of height and weight were taken.

The 1998 group were also asked more detailed questions:
  1. Have you ever tried to change your diet in order to change your weight?
  2. Do parents or family ever say you should eat more?
  3. How important is it to you that you like your weight?
  4. Do you ever think you should eat less?

30 girls were chosen from the 1998 sample because they scored higher than 20 on EAT-26. They were interviewed in detail about their eating behaviours and attitudes, their TV viewing habits and their opinions on “American TV”. The interviews were recorded and transcribed (typed).

Results

The girls’ actual weight didn’t vary much between the two cohorts. BMI (Body Mass Index) was average 24.5 in 1995 and 24.9 in 1998.

However, TV viewing increased: 41% came from a household with a television in 1995 compared to 71% in 1998. Becker carried out a Chi-Squared test and found the results to be significant at p≤0.001 level of probability. This means the increase in televisions found in this sample is 99.9% likely to be a statistical trend rather than chance variations.

The important difference was that only 12.7% reported an EAT-26 score over 20 in 1995, compared to 29.2% in 1998.

None of the girls reported using purging (induced vomiting) to control weight in 1995, but 11.3% reported this in 1998. Becker regards this as a particularly important finding.

Dieting is not a part of traditional Fijian culture, but in 1998 74% reported they felt too big or too fat and 69% reported they had dieted at some point. 83% felt that television was an influence in terms of body weight or body image and 40% linked losing weight to future success. (Unfortunately, these questions were not asked of the 1995 group.)

The qualitative data showed admiration for TV characters and a desire to be like them in terms of clothing, hairstyle and body shape. “I want their body… I want their size,” was one girl’s response.
When I look at the characters on TV, the way they act on TV and I just look at the body, the figure of that body, so I say, “look at them, they are thin and they all have this figure”, so I myself want to become like that, to become thin.
Conclusions

Dr Becker concludes that the characters on Western TV shows act as role models for Fijian girls. The girls feel pressure to imitate the slim appearances of actors and models, even if this means going against the Fijian culture that traditionally emphasises a fuller and more muscular figure.

Qualitative data suggested many of the girls did not understand the unreal nature of media images (eg that these actors lost weight specially for their roles and had personal trainers).

Becker believes TV has brought with it an interest in dieting that was not present before on Fiji; it has also brought with it the first symptoms of eating disorders into a society that previously had no incidence of anorexia or bulimia.
One could speculate that in the 20th century, television is another pathogen exporting Western images and values – Anne Becker

SOCIAL LEARNING EXPLAINS THE FINDINGS

Social Learning Theory claims that humans learn complex behaviour patterns through observing then imitating role models.

The Fijian schoolgirls have family members modelling healthy eating and larger body shapes. Normally, they would observe their parents eating habits and appearances and imitate them. This is because they identify with their parents and their paents have status.

However, the arrival of television brings with it a new set of Western role models who look very different from the girls' parents. These role models also have status (they are celebrities) and the girls identify with them too (many are youthful women who embody what the girls want to grow up to be).

Why would TV role models be more influential than real-life role models? Dr Becker points out that Fiji is going through an economic change from a farming economy (where the girls would expect to do the same work their parents and grandparents did) to a cash economy (where 80% of the the girls have ambitions to be things like airline hostesses). This means they identify with their parents less and see them as having less status. Several girls observed that TV encouraged Fijian children to be less polite towards their elders.

The idea that role models on TV can influence behaviour even when it is discouraged by parents is backed up by Bandura, Ross & Ross' 1963 study.

The Fijian schoolgirls were incredibly naive about television. According to Becker, the girls they interviewed believed that TV shows like E.R., and Friends were news stories and reality shows. They did not understand that the people they were seeing were actors or the difference between a scripted sitcom and real life. This means their cognitions (beliefs) about what they were watching might make them more likely to imitate. This is a good example of the way SLT incorporates Cognitive Psychology, unlike other learning theories.
although television is not the only source of idealised images of Western beauty available to Fijian adolescents … it is certainly the most accessible and most widely consumed medium and the only one introduced during the time frame of this study - Anne Becker

EVALUATING BECKER ET AL. AO3
GRAVE

Generalisability

The experiment involved two groups, both numbering in the 60s. This is a good size for a research sample, since one or two anomalous individuals (who, for example, watched excessive TV or no TV at all) should be “averaged out” by the rest.

However, it’s still a small sample to draw a conclusion from about the entirety of Fiji – which had a population of 850,000 in 1995. To draw conclusions about Fijian teens as a whole a larger sample would be more representative.

There may be problems generalising from Fijian girls to other cultures, because Fiji has such a strong culture of eating and distinctive notions of beauty and body shape.

The Fijian girls were incredibly naive about television. In contrast, British and American teen girls are very 'media savvy' and certainly don't mistake a hospital drama or soap opera for a news report. Becker herself points this out:
Generalization about the impact of television upon Fijians to other populations requires caution - Anne Becker
Reliability

This is a good example of a reliable study because it has standardised procedures and it was carefully documented. David Garner's EAT-26 questionnaire is widely-used in different cultures and is very reliable. The questionnaire was used with both groups and a score over 20 meant the same thing in 1995 as it did in 1998.  This is test-retest reliability.

There are some eating disorders the EAT-26 does not reliably detect. Binge-eating is an unhealthy eating behaviour but a binge-eater may still score low on EAT-26.

Becker's interviews were tape-recorded and transcribed so other researchers can read over what the girls said; this is inter-rater reliability.

Application

The main application of this study is for health workers. Fiji sets aide very little money for mental health, but Dr Becker’s study suggests that eating disorders and associated depression and suicide may become more common thee, so there is a need for more health spending.

Nicky Bryant, chief executive of the Eating Disorders Association, said Dr Becker's study had implications for other countries. Education may help young girls understand the unreal nature of slim models in the media. Studies like this may justify banning shows that feature particularly thin models, just as Size Zero fashion models were recently banned at Italian fashion shows.

Validity

However, only 29.2% - which is to say, 19 girls – scored dangerously high on the EAT-26 questionnaire. This is less than a third of the sample. 8 girls (12.7%) scored that high in 1995, despite Dr Becker’s claims that dieting and eating problems were “unheard of” in Fiji. While the numbers have gone up, the numbers suggest there were already problems with diet among schoolgirls and TV might not be the main factor.

This leads to another problem with validity: this is a natural experiment where the IV is not being manipulated. Lots of changes in Fijian society might have gone on (almost certainly were going on) alongside the introduction of television, so it’s difficult to be sure that TV is causing the changes in behaviour Dr Becker is reporting.

Another problem is the use of independent groups design. Although the two groups were the same age, with similar sizes and from the same schools, they would not have been exactly the same. It would have been better if Dr Becker had carried out a repeated measures design, looking at the same girls 3 years later to see how they had changed.
 
It’s important to note that the study did not find symptoms of real anorexia in either group of girls, nor evidence of weight loss. In fact, the average BMI for the girls in 1998 was 24.9 – and a BMI of 25.0 indicates being overweight! So perhaps these girls were concerned about being “too fat” because they were too fat!

Other researchers question whether eating disorders are really learned behaviours at all. Dr Hans Hoek (2005) led research on the Caribbean island of Curacao where fat is considered attractive. The incidence of anorexia on Curacao turned out to be exactly the same as Europe. This may mean there is a genetic cause for anorexia quite separate from Social Learning.

On a positive note, the study makes good use of qualitative data. As shown already, the study's quantitative data is ambivalent - it may or may not show what Becker concludes it shows. However, qualitative data from the interviews clearly gives the girls an opportunity to talk about their anxieties about weight and dieting and the pressure they feel to be like their TV role models.
Ethics

Dr Becker and her team were not specialists in eating disorders and were not competent to diagnose anorexia. Competency is important in research, because it would be unethical to give participants the impression they had a medical problem when really they just had some unhealthy dieting habits. Alarming and misleading participants like this would show a lack of social responsibility and a failure to mitigate risk.

To make the study ethical, Dr Becker looked instead for “worrying signs” like increased dieting or induced vomiting. The girls were not diagnosed with anorexia.

There may also be a problem with minimising harm to participants and their communities, because Becker's interviews might have provoked conflict between the girls and the older generation, by drawing attention to differences in their aspirations and outlooks (especially if the researchers gave the impression they approved of the girls' more modern outlooks).

EXEMPLAR ESSAY
An 8-mark essay on the Contemporary Study

Evaluate the contemporary study from the learning approach. (8 marks)
  • A 8-mark “evaluate” question awards 4 marks for AO1 (Describe) and 4 marks for AO3 (Evaluate). You need a conclusion for top band (7-8) marks.

Becker et al. compared groups of Fijian schoolgirls, one group when television was first introduced on the islands and another group 3 years later. They were looking for changes in dieting and body image.

Dr Becker’s study has a lot of application, because the Fijian government ought to devote more time and money to helping schoolgirls with depression and eating disorders brought on by TV viewing. Currently, Fiji spends little on this because eating disorders are very rare there.

However, it might not be possible to generalise the findings to other cultures, because Fiji is unusual with its focus on healthy eating and having a large size. Fijian girls might not be representative of women in other non-Western cultures.

The study might not be valid because it didn’t find any actual anorexia among the girls. Although the number of girls using vomiting to keep their weight down shot up from none to 11.1%, their average BMI was very high (24.9) which suggests they really should have been concerned about weight.

Becker's conclusions that eating disorders are learned may be wrong. Other studies have found anorexia in cultures where fatness is desirable, like the Curacao islanders. This suggests eating disorders may be genetic, not learned.

In conclusion, Dr Becker’s study backs up Bandura’s conclusions about role models influencing behaviour even on television. However, there fact that this was a natural experiment makes it hard to know what exactly is causing changes in eating behaviour.

  • Notice that for a 8-mark answer you don’t have to include everything Becker et al. did. I haven’t mentioned the sample sizes or the details of the EAT-26 test or the qualitative data from interviews. I haven’t described Becker’s conclusions. But I have tried to make the two halves – Description and Evaluation – evenly balanced.