Realiability and Validity Rosenburg Scale

Reliability and Validity of the Rosenberg Self Esteem Scale Texas A&M University Psychology 203 Introduction The purpose of our study was to evaluate the reliability and validity of the Rosenber Self-Esteem scale scores. Reliability is simply whether the measurement tool you are using measures something consistently. For example does the same test show the same results when administered repeatedly. Validity is the aspect of a measuring tool that signifies its measuring what it says it does.
A valid test measures what it says it’s measuring. However you can have a test that is reliable and not valid, for example if a teacher administers a spelling test that the student has to complete on the computer in a five minute time period. A student may get the same score over and over again, indicating it’s reliable, but it may not be valid because what if that student was slow at typing but knew how to spell every word. The test would not be valid because it’s not measuring just spelling as intended but also typing speed.
Both reliability and validity are crucial in and study because if the research instruments are not reliable and valid, then the results of your experiment will always be in question. Method All ninety-nine A&M students (N=99) that participated in the survey were in the same statistical writing course and were required to take the survey as part of the course curriculum. The survey consisted of 24 males and 75 females, with the average age of these participants being around 20 years old, ranging anywhere from 19 to 24 (M=20. 7, SD=. 997). the survey measured self-esteem by using ten items from the Rosenberg Self Esteem Scale. The students were scored by taking the mean of all their responses to the ten questions. 10 different statements that the participant rated on a scale of 1 to 7, (1 being “disagree strongly ” and 7 being “agree strongly”), the higher the score the higher the individuals self-esteem. The scale consisted of 5 revers coded items; an example of a normal item is “on the whole I am satisfied with myself”.

A reverse coded item is an opposite statement for example is “I feel I do not have much to be proud of”. Results The 10 item subscale from the Rosenberg Self-Esteem scale appeared to have good internal consistency, (? = . 90). All the survey question appeared to be worth keeping: in fact the deletion of any of the questions would have resulted in a lower alpha score. All 99 participants (N=99) responses to the 10 item survey were averaged together and the self-esteem score for all the participants had ranged from 2. 6-7 with a mean of 5. 4 (M=5. 4, SD=1. 08).
The frequency distribution of the average self-esteem scores, illustrated on the histogram, portrays a distribution that is pretty normal but somewhat negatively. This negative distributions means that the participants more frequently had high self-esteem scores than low scores. In fact no participants had a self-esteem score lower than 2. 6. Discussion To ensure that the survey instrument used in this study was actually measuring self-esteem we took a measure of Cronbachs alpha ( or ? ), which is a special measure of reliability known as internal consistency.
The internal consistency reliability of survey instruments (e. g. Rosenberg Self-Esteem scale), is a measure of reliability of different survey items intended to measure the same characteristic, in this case self-esteem. For example, there were 10 different questions related to self-esteem level. Each question implies a response scale from 1 to 7. Responses from the group of the 99 respondents have been obtained. In reality, answers to different questions vary for each particular respondent, although the items are intended to measure the same aspect or quantity.
The smaller this variability (or stronger the correlation), the greater the internal consistency reliability of this survey instrument. So in this survey the self-esteem scale was found to be highly reliable (10 items; ?? = . 90). We can safely assume that our survey items reliably measure self-esteem levels. One other way we might have assessed reliability would have been to get all the participants to take another Self-Esteem survey of similar design but with different questions.
In social sciences in particular making sure that your research has construct validity is very important. Construct validity has traditionally been defined as the experimental demonstration that a test is measuring the construct it claims to be measuring, in laymen’s terms; does the measuring tool actually measure the theory under examination.? “Am I actually measuring what I think I am measuring? In our example construct validity would be how well dos our observational tool (Rosenberg self-esteem scale) assess one dimension of self-esteem.
We could of inter-correlated measures of depression with the measures of the revers coded items (SE3,SE5,SE8,SE9,SE10), because depression and the revers coded items have similar underlying constructs. Finally the low correlation between height and self-esteem in our study strengthens our construct validity. I believe it strengthens the overall resolve because common theories of self-esteem have no reliable and valid evidence correlating the two, so we can assume that construct between the two would be low, and if our study showed a strong relationship it would be casue for questions.