Hey, sorry to repost, but the other came out really difficult to read so i have delete tables etc. Could someone please proof-read my work. I am dyslexia and find proof-checking my work near impossible. It is long but will probably only be stupid mistakes.
***
Introduction
Due to social and media pressures the ideal body shape is becoming thinner and this contributes to eating disorders in both males and females Harrison & Cantor (1997).
Anorexia nervosa is defined by the DSM-IV (American Psychiatric Association, 1994) as being 15% under recommended weight for height, age and gender. The patient also has to have a distorted body view and a drive to become thin.
Gordon, (2000) states the ratio of females suffering from an eating disorder compared to males at 9-1. He recognises a gender difference in eating disorders but urges people to acknowledge the existence of the male sufferers. He describes a male patient had visiting a doctor, who had replied to the patients anorexia as “a problem that only happens to girls”. He claims that being labelled as suffering with a “female” disorder can lead to the male feeling ashamed and emasculated and therefore more hesitant to get help, and for this reason some believe that male anorexia is largely unreported. Gordon claims this feeling of shame is enhanced when even female anorexics are judgemental at group therapy.
This negative attitude towards male anorexia causes health risks. It may result in professionals and the males themselves not recognising their early symptoms and therefore may develop chronic anorexia nervosa before they are diagnosed. (Margo, 1987, cited in BMA publication 2000)
Whitney et al (2005) studies individual experiences of people who carers of anorexic people. The females’ reports were more emotional than the males who appeared to be less attached with a less sympathetic viewpoint, suggesting males have a more negative attitude towards anorexia nervosa than females do.
Support that males have unhealthy eating patterns in addition to females comes from Forman-Hoffman (2004) who conducted a study involving 9,118 adolescents, 7.3% of females and 3.1% of males reported unhealthy eating patterns over the previous week.
However, theory correlates with the attitudes as many eating disorder theories are specific to women. For example some psychoanalysists argue that anorexia nervosa is developed during whilst females experience separation anxiety from their mothers. (Darvell, BMA, 2000)
Literature on male anorexia nervosa is limited and there does not appear to be any prior research into the difference between attitudes towards male anorexia nervosa and gender. Furthermore, there appears to be a fair amount of ignorance when it comes to male anorexia nervosa, present in medical professionals as well as psychological theories not even accepting its existence might be suggesting that additional education about male anorexia needs to be put into place.
To clarify an attitude is a negative, positive or neural feeling towards an “attitude object”, which does not have to be a physical object; it could be an event for example. Attitudes are formed by affective and cognitive responses and behavioural intentions. The following study only examines explicit attitudes, which are attitudes which are self-reported, which can be easily measured on a likert scale.
The aim of this study is to investigate the different attitudes towards male anorexia nervosa between males and females by use of a questionnaire will remain reliable over time.
The hypothesis for this study is gender will affect attitudes towards males suffering from anorexia nervosa.
Method
50 male and 50 female students were selected by opportunity sampling to be participants for the study. Each participant was asked to fill out two questionnaires a week apart. The males were either part of a university football club or an engineering society, while the females were psychology students.
A 5-point likert scale was used.. The questionnaire used 18 affective, 18 behavioural and 14 cognitive. 25 of the questions were testing for a negative attitude, and 25 for a positive attitude.
The first questionnaire conducted had 50 questions but after the analysis had been carried out questions were removed from the questionnaire. Below is a table showing the removed items and why they were removed.
The overall design was a survey. The independent variables (IVs) of this study were gender of the participant and which questionnaire was completed. Gender is a between subjects IV, which has two levels; male and female. Which questionnaire completed is within subject IV, as each participant completed both questionnaires. This IV has two levels, the first questionnaire and the second questionnaire. The dependent variable (DV) is the scores on the questionnaire. The predictor variable (PV) in this study was the scores of the first questionnaire, while the criterion variable (CV) in this study was the scores of the second questionnaire.
The participants’ details were anonymous and were only identified by their mothers’ maiden name and consent was gained from all participants involved who were briefed beforehand and understood that they could withdraw at any time.
First a questionnaire was constructed by brainstorming statements indicated positive and negative attitudes towards male anorexia nervosa and a mixture of affective, behavioural and cognitive, (18,18,14 respectively) with 25 negative attitude questions and 25 positive attitude was deliberately used. Then the questionnaire was typed up with a 5-point likert scale, with the options of “1. Strongly Disagree”, “2. Disagree”, “3. Neither Agree Nor Disagree” “4. Agree” “5. Strongly Agree” beneath each question. A brief and a form of keeping the participants details anonymous by using their mother’s maiden name, was also constructed and then potential participants were approached and asked to partake in the study. Once 50 male students and 50 female students had completed the questionnaire the results were coded into numbers as shown above. Then the negative attitude questions were recoded so that 5=1, 4=2, 3=3, 2=4 and 1=5, and then the questionnaires were both qualitatively and quantitatively analysed.
Qualitative analysis was done by looking at the questionnaires’ results and seeing if there was any reason why any particular question shouldn’t be there. For example, if most people selected “neither agree nor disagree” then the question clearly did not serve its purpose. Quantitative analysis involved looking at the mean, and deleting any items which lay considerably outside 2.5-3.5 margins. Any items with a significantly larger standard deviation or lower correlation figures were also deleted. Then with the questions deleted new quantitative analysis was conducted to check that the results were significant. Then the questionnaire was handed out to the same participants to complete. The second set of data was then reanalysed and a t-test was carried out to compare the totals of each participant for either questionnaire to check the questionnaire was reliable over time, and Levene’s test was perform to check for a homogeneity of variance. None of the participants where told we were comparing male and female attitudes until after the questionnaires were completed at which point they could disagree for their data to be used.
Results
The item means statistics are as would be expected when testing two opposing attitudes on a 5-point likert scale. From table 2 it is clear that all the means and standard deviations are similar. With the standard deviation being around 1. The item means are as they should be, with the mean around 3, (M = 3.137) with a few deviations, (Range = 1.290, Variance = 0.124,) no items having a mean of much less than 2.5, (Min = 2.470), or much higher than three point five, (Max = 3.760). The item variance is fairly low (M=1.011) and deviates little (Range = 0.634, Variance = 0.33)
These show similar results to those found in the 1st questionnaire. Standard deviation are around 1. Means are around 3, (M = 3.128) with a few deviations, (Range = 1.320, Variance = 0.136,) These very similar results would suggest that the questionnaire had consistency over time; however this is tested for and discussed further later.
Reliability analysis of 1st and 2nd questionnaires
A reliability analysis of the items of the first questionnaire was performed to explore to what extent the scale was internally consistent. The results showed strong internal consistency in both questionnaires (Cronbach's Alpha = 0.891). The result suggest that the scale is not unidimensional (Inter-Item Correlation = 0.289) Similar findings were found with the reliability analysis of the second questionnaire, there was a similar strong internal consistency (Cronbrach’s Alpha = 0.882) and the scale still appeared to be multidimensional (Inter-Item Correlation = 0.270)
Test-Retest Reliability
The results above show that both tests are internally consistent but to discover if the test was consistent over time a test-retest correlational analysis was conducted, scores from the same individuals, on the same items on both questionnaires should be similar to indicate reliability over time. Each participants’ total score was calculated for questionnaire one and questionnaire two and then the relationship between the two variables was examined. A significant strong correlation between scores on the two questionnaires was found (R = 0.949, p = 0.001). These results indicate that the test was reliable over time. The strong correlation between the two tests is illustrated below in a scatter-graph.
The hypothesis for this study was 1-tailed. The females scored a mean of 70.04 while males scored a mean of 55.06. An independent t-test was conducted to explore the influence gender had on attitudes towards male anorexia. A Levene’s test of homogeneity of variance was carried out and found to be non-significant (p = 0.183) and therefore it can be assumed that there are no significant differences between variances. Gender was shown to have a significant effect on attitudes towards male anorexia, (df = 98, t = 8.045, p = 0.001).
Discussion
To summarize the results, internal consistency and consistency over time were present in the questionnaires; homogeneity of variance was assumed an independent t-test yielded significant results indicating that there is a difference between male and female attitudes towards male anorexia with males holding the negative viewpoint, while the average female has a more positive attitude towards male anorexics. This is supports the hypothesis that gender affects attitudes towards male anorexic.
Although these findings are highly significant it would be beneficial to conduct research where the scale was unidimensional as opposed to multidimensional.
These results support that Gordon’s viewpoint (2000) was correct that males suffering with anorexia nervosa are stigmatised, however Gordon also expected females to have negative attitudes towards anorexic males which this study does not find supportive evidence for. To ensure that it is not just female anorexics that stigmatise male anorexics a study could be conducted to investigate the difference between female who are suffering with anorexia nervosa and females who are not.
Negative attitudes towards male anorexia may be due to males being undereducated about anorexia as the condition is mainly associated with females. These results could be the consequence of a lack of media coverage. Female bias in academic publications about male anorexia may also contribute (Darvell, 2000). According to Margo, (1987) this will indirectly cause anorexic males health to become worse.
All the participants where students, with no mature students included, which served as a control, however further research to see if the results can be generalised would be worthwhile.
All the participants completed the questionnaires in similar conditions and the use of a questionnaire meant all the questions were standardised. Opportunity sampling was a suitable method for this study due to limitations on time, and the results were highly significant, but if a random sampling method was used for similar research in the future it would reduce possible criticisms. Because none of the participants where told we were comparing male and female attitudes until after the questionnaires were completed, the likelihood of demand characteristics was minimised.
To conclude, the results show a significant difference between male and female attitudes towards anorexia. Males showed a more negative attitude. The questionnaires internal consistency and reliability over time, along with homogeneity of variance of the results can all be assumed. The negative attitudes theorists suggest could be affecting the sufferers’ physical and mental health (Margo, 1980; Gordon, 2000) so it is a vitally important area of study to as evidence will in itself raise awareness and encourage other means of doing so