Self-esteem can be defined as an individual’s sense of worth and the standards by which they judge themselves (Branden, 1969) and is a method of self-evaluating oneself in either a positive or negative manner (Baron & Byrne, 1991). Much research has been carried out to discover if levels of self-esteem differ depending on gender. Many of these studies have shown that girls will have significantly lower self-esteem than boys (Skaalvik, 1986; Harper & Marshall, 1991; Kling et al. , 1999; Robins et al, 2002; Joshi and Srivastava, 2009; Rahmani, 2011).
Other research, however, has found that the evidence for lower levels of self-esteem in girls is ‘surprisingly thin’ (Renshaw, 1990, p. 18). Watkins et al. (1997) contradicts these findings as he concluded that girls had higher levels of self-esteem than boys. Robins and Trzesniewski (2005) found that self-esteem in both genders was relatively high in young children but as the children moved into adolescence, boys developed higher levels of self-esteem than girls. So it would appear that the age at which the level of self-esteem is measured affects the results obtained.
Another reason for contradicting results could be the use of different tests to measure the levels of self-esteem. Some researchers (Skaalvik, 1986; Maehr & Nicholls, 1980) suggest that some of the tests are biased as they encourage the participant to show possible exaggerated feelings of confidence and pride, traits which are often more associated with male bravura. Girls on the other hand may be more modest in their answers (Skaalvik, 1986; Eccles et al. , 1984). Many other studies support this idea as they have found that the elements which make up self-esteem are gender specific (Knox et al.
, 1998; Dusek and Flaherty, 1981). Josephs et al. (1992) found that men related self-esteem to having ‘superior abilities’ whereas interconnectedness related to higher levels of self-esteem for women. Indeed Block and Robins (1993) found that an adolescent boy’s self-esteem was more self-oriented and a girl would generally be more interpersonally oriented. Research Question Previous research would therefore suggest that levels of self-esteem can be affected by age range and the reliability of the test used to measure it.
This research was therefore limited to examining the relationship between self-esteem and gender in primary school pupils in years 4 and 7. Pupils’ self-esteem was measured using a subscale of Harters’ (1985) Self Perception Profile for Children as it was designed specifically to measure self-worth of children aged 8-12. This scale splits into six different subscales; Athletic Competence, Physical Appearance, Social Acceptance, Scholastic Competence, Behavioural Conduct and overall Global Self-worth (Harter, 1985).
As it measures different areas of self-esteem in which one gender may score more highly than the other, e. g. boys may score more highly in athletics and girls in behavioural conduct, it is considered to be a reliable scale which is not gender biased. The aim of the study will be to test the null hypothesis that there is no difference between the levels of self-esteem in boys and girls. The alternative hypothesis, however, will be that there is a difference between self-esteem of boys and girls. Method Sample
The research was based on data from150 pupils, which was collected as part of a larger study involving 1080 children. The data collected measured the educational achievements and well-being of Year 4 and Year 7 pupils. Before taking part in the study informed consent was sought from the pupils’ parents/guardians and pupils were made aware that they could withdraw from the study at any point. A pilot study was then completed with a sample of 60 Year 4 and Year 7 pupils to ensure that the questionnaire was worded appropriately, was not overly time consuming and provided all the necessary information.
The questionnaire was then completed in the pupils’ own classroom, as it was a familiar and safe environment, and the questions were read aloud before the pupils would make their response to ensure different levels of literacy wouldn’t affect pupils taking part. Measures and Variables There were two variables within this research. The first was the gender of the pupils which created a nominal variable. The second was the pupils level of self-esteem which was measured using a subscale of Harter’s Self Perception Profile for Children which created a scale variable.
Some of the reasons for choosing this scale have already been stated but previous research has also reported it to be highly reliable as the Cronbach’s alphas range for internal consistency falls between 0. 73-0. 81 and the intra-class test re-test correlation is above 0. 84. Analysis This data was then entered and analysed using SPSS software. As the gender of the pupils taking part created a nominal variable the frequencies were calculated and summarised in a table.
The level of pupils’ self-esteem was recorded using a histogram as it is an appropriate method of recording scale data and can quickly show if the distribution is normal or skewed. As the results produced a normal distribution both the mean and standard deviation was calculated. The percentages of pupils falling within plus or minus 1 and 2 standard deviations from the mean were also recorded. The differences between the mean self-esteem scores for boys and girls were then calculated and displayed using a histogram and the mean and standard deviation were calculated.
The range of scores achieved by boys and girls was then calculated plus or minus 1 and 2 standard deviations of the mean and the findings were displayed in a table. Finally the Independent Samples T-test was then performed to assess if the results could be just a sampling error or if they would be representative of the population as a whole. This was the most appropriate statistical test as it assessed the relationship between the two different groups of participants, as it is a bivariate statistical technique, and it could be used to compare nominal and scale variables.
Findings The findings of the research are displayed below, firstly looking at the descriptive statistics and then progressing on to the inferential statistics. Descriptive statistics From the frequency table below (Fig. 1) we can see that the sample consisted of 150 participants in total. Of this, 72 (48%) participants were male and 78 (52%) were female. Therefore the sample used had slightly higher percentage of female participants than male.