Aim
“To find out if self esteem is positively effected by academic success among students aged 16-18”
Hypothesis
The results of this experiment will show that there is a significant positive correlation between males and females who perform well academically and those who gain a high point score on a self-esteem Questionnaire.
Null Hypothesis
Students who have performed well academically will not necessarily gain a higher point score on a self-esteem questionnaire than students who have had less academic success. Any results are due to chance.
The hypotheses are 2 tailed because although it is expected that there will be a ‘direction’ in results, the direction in which this correlation will be (positive or negative) is not sure. As show in previous studies such as James’ (1890), variables such as pretences have an influence on results which cannot be measured as there is no background information on each individual participant.
Method
Design
The study that is going to be conducted is a field experiment, doing a field experiment means there is a reasonable amount of control over vairables The investigation will be carried out in the school common room or in classrooms, meaning it will be conducted in a natural environment for students. There is high ecological validity in this type of experiment. The academic data collected is from GCSE results gained at Rastrick high school, and the self-esteem ratings are gained from a revised version of Rosenberg’s self esteem scale. The questionnaires will be filled out in Rastrick High School as it is a natural, everyday environment for the participants, this should mean that the participants feel comfortable and at home, giving high ecological validity. The study being conducted is a correlation investigation. To gain statistical data Spearmans Rho statistical testing will be used.
Variables
IV - the gender of the participants.
DV - the self-esteem score generated by questionnaire.
Confounding Variables
- Interruptions
- Experimenter Bias
- Personal Background
- Noise Level
- Time of Day
- Temperature
- Age of participants
- Background of each participant
- Pretences held by each participant
As the study is a field experiment variables such as temperature and noise level are part and parcel of the everyday environment that the participants are used to. The age of participants is a big confounding variable as the age difference between any two people in year 12 at Rastrick High school could be almost a year, i.e. one participant could be born in September 1986 and another in August 1987. This age difference may have a large effect on self-esteem as their ideas of ‘success’ and what they want out of life may be different. To try and control this the study will have 15 participants from each gender
James (1890) states that an intervening variable in a study such as this is personal expectations. Other aspects, which may cause an experimental error, may be the environments in which the participants fill in the questionnaire.
Participants
The target population is year 12 students at Rastrick High school who completed their GCSE examinations in the summer of 2003. This target population has been chosen because they have had recent important examination results over a broad range of subjects and will all be within the 16-18 age range that is being investigated. Random sampling will be used by getting a copy of the entire year 12 register, splitting up the register into boys and girls, then giving each a number from 00 upwards. Both sets of numbers will be taken and placed each in a separate container and randomly select 15 from each. The numbers will then be matched to the ones from the register and the individual asked to fill out the questionnaire.
Apparatus and Materials
- Questionnaire
- Common Room/Classroom desk
- Pencil
- Rubber
Procedure
The participant who has been randomly picked will be approached. They will be asked if they have the time and if it is convenient for them to complete a questionnaire. They will be briefed (appendix 1) and given standardised instructions on how to fill out the questionnaire (appendix 2) and allowed to fill in the questionnaire without distraction. When they have finished They will ve debriefed (appendix 3) and thanked for their time.
Ethical Issues
- As the participants are over 16 they do not require a consent letter, however I will ensure that they fully understand what the study is about and get full verbal consent.
- I will remind participants that they may withdraw at any time both in the briefing and de-breifing.
- I will ensure that I avoid deception of the participants.
- I will ensure participants are correctly de-briefed
- I will ensure participants know they may withdraw from the study at any time.
- I will respect their right to privacy
- I will try to avoid the participant suffering physiological and psychological discomfort by creating a relaxed environment with good communication. I am aiming for my participants to be in the same psychological state after the investigation as they were before.
- The data will be kept confidential at all times
- In my results, names will not be used.
Consent
Informed consent is gained from Rastrick 6th form college, as the participants are over 16 informed consent from the participants parents is not required. The participants will be informed of all the aspects of this investigation that could influence willingness to participate.
Deception
No information will be withheld in this investigation, and the participants are not tricked or mislead in any way. Participants will still be fully briefed before the study and de-briefed after.
Withdrawal
Participants will be informed that they many withdraw from the study at any time, and that if they choose to withdraw then their results will be terminated and not used in any way for results.
Briefing
Hello I am Paul from the upper 6th. For my psychology coursework I am looking into self esteem among year 12 students. I would like you to fill out a simple questionnaire, if there is anything you don’t understand please ask. You may withdraw from the study at any time. The result of your questionnaire will remain confidential, and should you wish not to write your name on the questionnaire any remain anonymous that is fine. If you would like to view the results I will contact you when they become available.
De-Briefing
Thanks for taking part. Basically I am looking into the relationship between academic success and self esteem. Do you have any questions? Should you wish to speak to me or Mrs Hall at a later date please do not hesitate to ask.
Standardised Instructions
I am going to give you a questionnaire, if you would like to write your name, do so at the top, then please answer the questions in turn, take as long as you would like, there is no rush. Should you make a mistake there is a rubber available to change your answer. When you have finished I will collect the questionnaire back off you.
Graphs
Summary of Data
To gain statistical data I have used Spearman’s Rho statistical testing. To decide which type of statistical data to use I had to compare my experiment with different criteria. My the data I gain from my experiment will be ordinal and that distribution of data is non-parametric. As the data is non parametric my data may be more open to type I or type II errors.
Correlation can be measured as decimals from –1 to 1, and it can be positive or negative. Negative correlation is measured between –1 and 0, and positive from 0 to 1. The strength of correlation (positive or negative) is judged by how far away the decimal result is away from 0. For example +/- 0.1 would represent a weak positive/negative correlation and +/- 0.9 would be a much stronger correlation. A correlation of zero would represent no correlation.
The results show that girls had a positive correlation of 0.221 and boys had one of 0.846. This shows that the boys had a much stronger positive correlation. Therefore I can reject my null hypothesis as the results of my investigation show a positive correlation between self-esteem and academic success.
Results
Summary Tables
Academic Scores
Self Esteem scores
Graphs of Summary Tables
Discussion
The results of my Spearman’s Rank show that, for both boys and girls that took part in my study, there is a positive correlation between academic success and self-esteem.
By entering the raw results I found by doing my study (appendix 4) into the Spearman’s rank formula I was able to find that girls had a positive correlation of 0.221 and boys gained 0.836 where +1 is perfect positive correlation and –1 is perfect negative correlation. This shows that boys had a substantially more positive correlation than girls, and therefore relates better with my original alternative hypothesis – which was that students who perform well academically have higher self-esteem than those who perform poorly. For my study on boys alone I believe I can reject my null hypothesis on the grounds that there does seem to be a substantial positive correlation between academic success and self-esteem. However on my study of girls I do not think that a correlation of 0.221 is a strong enough one to describe as substantial and therefore my null hypothesis cannot be rejected all-together.
My summary tables show that girls have a higher mean result than boys in academic scores, and that the girls scores are more consistent than the boys as the standard deviation is lower. This goes along with the popular idea that girls usually perform better academically than boys. However is self-esteem boys gained a marginally higher mean score and a lower variance in results.
There are however limitations with my study that (according to James – 1890) would have a profound effect on accepting or rejecting my hypotheses. In my study I have no background information on any of the participants or any indication of what sort of psychological state they may be in. For a self-esteem study such as this one a detailed initial questionnaire would have to be filled in by participants before other results were gathered. This would be a very tricky process to try and gain personal information and ethical issues would have to be watched closely to ensure that no psychological distress was put upon participants. I also evidently have no evidence of positive or negative relationships in early life, which according to Cooper-smith (1967) and Growe (1980) are vital in the self-esteem a person has later on. These ideas also link high self esteem to good task performance, which could suggest that it is self esteem initially which assists in gaining academic success, rather than the other way around which I have been investigating.
Expectations are another aspect of a persons psychological state that I have not been able to include in my estimations or predictions of my results. Each participant is going to have different aspirations and ideas of what denotes ‘success’ when it comes to academic studies, some peoples expectations are bound to be higher than others.
The age group I have investigated may also effect the validity of the study. Stereotypically 16-17 year olds have ‘raging’ hormones and varying emotions which could swing self-esteem either way at any given time. It could be said however that there is not a real ‘ideal’ time to do the study, as post A-levels students are more focused on the specified career path they are aiming to take and academic studies in a number of subject is not as important.
I think over a large enough population the results of experiments such as the one I conducted and many like it may be generalised, as my results show and those of other investigations mentioned in my introduction, academic success can effect self esteem positively and over a large number of people a positive correlation is likely to occur. However, individually I do not think its possible to predict self esteem purely on academic success as each entity would have different backgrounds and pretences.
The variation in results between males and females could be down to gender differences. Girls may take more notice of outside influences such as their appearance or other social issues than boys do. If a future studies were to be done all these limitations would have to be taken into account if the investigation was to gain high validity. For future investigations to gain any ‘concrete’ results I think they would have to be more specific in order to control confounding variables. For example a study could be done in which all the participants had experienced a particular ‘trauma’ in their life such as divorce or a death of close relative. The results of these experiments may show how pretences effect self-esteem in young adults, and by testing on different age groups it could be possible to investigate whether pretences become less apparent in later life or if they get stronger.
Conclusion
The findings from my study have shown that, in boys especially, there seems to be a positive correlation between students who perform well academically and those with high self-esteem. However I believe that self-esteem is not something that can be measured so easily, I consider outside influences to have a huge say in self-esteem and the behaviour that it entails, influences that are extremely hard to measure and detect from simple questionnaires.
Abstract
Studies in the past have related self-esteem to virtually every other psychological concept or domain, including personality (e.g. shyness), behavioural (e.g. task performance), cognitive (e.g. attribution bias) and clinical concepts (e.g. anxiety or depression). James (1890) said that academic success did have an effect on self-esteem, my investigation looks into the effect academic success has on self-esteem in boys and girls. It will be based on an adapted study of Rosenberg’s self-esteem scale. I will also be re-testing James’ hypothesis as the study was conducted over 100 years ago
In my study I am aiming to find out if self-esteem is positively effected by academic success among students aged 16-18. I predict that there will be a significant positive correlation between males and females who perform well academically and those who gain a high point score on a self-esteem Questionnaire.
I will be conducting a field experiment and the experimental design I am carrying out is a correlation one.
My IV is the gender of the participants.
My DV is the self-esteem score generated by questionnaire.
My results showed that while a positive correlation was found (in boys more than girls) outside influences may prove to be huge limitations on my results.
I conclude that my hypothesis may have been largely correct, however I cannot definitely say that my results are valid and therefore acceptable in such a study. The investigation has shown that the way self-esteem effects behaviour is hard to prove and accurately put into statistics.
