Purpose Statement
This study will be focused on understanding the relationship or causality between anabolic steroid use and depression among high school teenagers between the ages of 9–12. Many teenagers today are surrounded by many adverse conditions which eventually negatively affect their behaviors and attitudes. It is no surprise therefore that many teens today are more depressed than they should be at their age (Schoon, 2006, p. 132). The kind of social adversities affecting teens today have been researched for many years and contemporary researchers have not shied away from noting that such adversities are bound to affect the development of teens in cognitive and psychological ways (Schoon, 2006, p. 132).
There are many factors affecting the degree or severity of various adverse effects which teens are faced with. Commonly, the nature and frequency of such adversities are known determinants of the severity of psychological impairment among teenagers (Schoon, 2006, p. 132). These factors withstanding, this study will seek to establish the relationship between such adversities and the propensity of a teenager developing depression symptoms. Comparatively, the study will establish whether adversity increases the propensity of depression in solitude or does it has to be coupled with steroid use to cause depression?
Research Question
Does anabolic steroid use cause depression in teenagers who never had a history of depression before?
Null Hypothesis
There is a high probability of teens taking steroids scoring highly on the depression scale.
Alternative Hypothesis
There is a significant correlation between students scoring highly on steroid use and those scoring highly on the depression scale.
Measures
The measures to be employed in this study will include index construction and scaling methods. Indexes and measuring instruments are used to measure specific grounds of commonality between human behavioral tendencies (Vaske, 1990, p. 49). The specific tools of scale measurements to be used incorporate the Likert scale and the Guttmann scale. The Likert scale will be used to measure respondents’ attitudes after a random compilation of the respondent’s views is administered. Later, a common score will be given to each respondent. In this manner, the researcher will be in a position to determine which of the two indicators is strongest, and then later, the indicators can be narrowed down to the most basic and relevant. Afterward, the scale can be administered again to gauge the strength of each indicator (Frankfort-Nachmias, 2007, pp. 414-415). The Guttmann scale will also be used as an empirical test to determine the unidimensionality which may be experienced when constructing the scale in the first place. The biggest advantage with this type of scale is that we can evaluate the number of responses on a given score by matching or comparing one score with another (Frankfort-Nachmias, 2007, pp. 422-431). This will ultimately establish the conformity of respondents’ views.
As mentioned earlier, the respondents will be high schools students and the measurements discussed will be undertaken on a group of fifty respondents. These types of measurements will be the most appropriate when exploring teen depression caused by exposure to steroid use and adverse environmental effects. In the past, these techniques were used by many social scientists to comprehensively understand gender relationships, power relations, freedom, intelligence and such like human variables and so they are equally appropriate for this study (Frankfort-Nachmias, 2007, pp. 412-413).
Evaluation of Measurement Choices
The indexes and scaling methods to be used will be important in this study because they are composite indexes and incorporate more than one variable (which is a very useful tool in this study because it will explore two variables: adverse environmental effects and steroid use) (Frankfort-Nachmias, 2007, pp. 414-415). This appropriateness can be equated to the use of the Likert and Guttmann scales in measuring the socioeconomic status of individuals through analyzing various indicators such as income, education and general occupation (Frankfort-Nachmias, 2007, pp. 422-431).
The Indexes and scales to be used are also quite useful in this study because they are very practical and have the potential of representing a number of variables in a single score (Frankfort-Nachmias, 2007, pp. 422-431). This is a very important attribute considering the complexities experienced when trying to represent more than one variable in a single study will be eliminated. Also, many times, indexes and scales have been confirmed to score highly in the precision scale when it comes to analyzing the variables in question (Frankfort-Nachmias, 2007, pp. 422-431). Furthermore, these instruments have been confirmed to have a high capability of stabilizing the measurement scale in the first place (Frankfort-Nachmias, 2007, pp. 422-431). In other words, the Likert and Guttmann scale to be used are very consistent in measuring causality and such like relationships. Nonetheless, the ultimate advantage of using these scales is their ability to respond or answer questions touching on more than one element (Frankfort-Nachmias, 2007, pp. 422-431). This is especially noted in the Guttmann scale. This advantage is generally approved in most spheres of research. For example; often, students would rather be examined by a multi-sectional grade in exams because it makes more sense to have a number of choices or options to a given question since there is no universal answer to most questions (Frankfort-Nachmias, 2007, pp. 422-431). The use of two variables in the study is, therefore, an effective method of analysis.
However, most importantly, the Likert and Guttmann scales will be important to this study because, from various points of view, we can analyze more than one element affecting depression (Frankfort-Nachmias, 2007, pp. 422-431). This comes about because sometimes it is increasingly difficult to have a general analysis of a given causal relationship without analyzing the existing underlying factors behind such relationships. Most of the time, such underlying factors are more than one. For instance, Frankfort-Nachmias (2007, pp. 425) gives an example that it may be unfair for a student to be graded on one question in a given subject area because it may lead to an unfair judgment of the student’s cognitive skills due to generalization. This is true because if a student is given a single question and he/she makes an error while doing it, the professor or instructor may wrongly grade the student based on the given error alone. However, if multiple questions were offered, the student can have an opportunity to truly express his/her understanding of the subject. From such ambiguities, the indexing and scaling techniques become very important in this study because they have a high degree of precision than most data analysis instruments. In this regard, the scales to be used will provide a high degree of unidimensionality and therefore many people or researchers can use a number of techniques to come up with specific conclusions on their studies. In addition, through the use of indexes and scales, we can be able to rank steroid use or related adverse effects based on their ability to drive teenagers into depression (or not). Most data analysis techniques don’t have the ability to show interval level scale which will be achievable in this study when analyzing nominal and ordinal data to be used in the study (Frankfort-Nachmias, 2007, pp. 422-431).
Since we will be able to rank the variables in the study, we will be using the norm referencing method. Later, the scores will be tabulated through statistical techniques for future analysis. Practically, this will determine whether steroid use in isolation causes teen depression or whether it ought to be coupled with the presence of adverse environmental effects to result in the same outcome.
Level of Measurement
The level of measurement to be incorporated in this study will primarily depend on the scales to be used. Considering the fact that Likert and Guttmann scales will be used in this study, the level of measurement of the study will be strictly limited to the level of measurement the Likert scale and Guttmann scale have. One on hand, the Guttmann scale will analyze elements that agree with each other (or in other words, correlated elements of steroid use and adverse environmental effects on teen attitude and behavior). From this understanding, the Guttmann scale will analyze acceptable positions and extreme positions held by the participants. The Likert scale on the other hand will be used to evaluate the degree of agreement or disagreement of the respondent’s view and not necessarily the position held by the respondents in general (as measured by the Guttmann scale) (Frankfort-Nachmias, 2007, pp. 422-431). The responses may range, say from “strongly agree” to “strongly disagree” and therefore the degree of respondents’’ view can be easily noted.
Reliability of Test selections
For the Likert scale, the reliability test to be undertaken will not be much different from the Guttmann scale. The variables to be analyzed will however be taken in large amounts to suit two scales and the split-half reliability test will be administered (Frankfort-Nachmias, 2007, pp. 422-431). This will affirm its reliability. With regards to the Guttmann scale, its reliability will be tested when the same tests are carried out and analyzed on a different type of scale (Plooy, 1995, p. 80).
Validity of Test Selections
The validity of the Guttmann scale can be checked by first checking for high coefficients of scalability (Osgood, 1971, p. 194). If the coefficients for scalability are high, the Guttmann scale is valid but if the coefficients for scalability are low, the scale may be invalid. Another method of checking the validity of the Guttmann scale comes from the unidimensional nature of the Guttmann scale itself. This attribute increases its validity. With regards to the Likert scale, its validity can be checked by running the test on a group whose attitudes are already widely known (Erwin, 2001, p. 53). If the results obtained from the scale do not match the expected results, then the probability of the scale being valid is not true. For instance, this study is social in nature and therefore there already are conventional perceptions regarding adverse environmental conditions and teen depression or steroid use and teen depression. The results obtained from the Likert scale should therefore be analyzed with such generally expected attitudes.
Evaluation of Tests
The Likert scale is quite useful when measuring the attitudes and feelings of respondents because it is ordinal and enables researchers to rank various manifestations of attitudes and behaviors (Denzin, 2006, p. 164). However, unlike the Guttmann scale, the Likert scale cannot explain the differences between the attitudes or behaviors observed (Kumar, 2010, p. 108). For instance, in this study, the Likert scale will be able to expose various measures of depression but cannot explain the differences between the measures. It can only rank them. However, through its wide range of responses, the scale will be advantageous because it has a high degree of precision. Nonetheless, the extensive scale of responses can be deemed a little discriminating as is observed by most researchers when compared to the Guttmann scale (Lodico, 2006, p. 108). Also, since the Likert scale incorporates about 30 statements (on average) to encompass the possible responses of the participants, the chances of vague statements being chosen are unavoidable. Nonetheless, the Likert scale lacks a “judging gap” in its analysis (Kumar, 2010, p. 108).
The Guttmann scale though quite elaborate is probably more difficult to prepare when compared to the Likert scale. This is because it takes a lot of time analyzing the responses and coming up with related questions to give the respondents an overall score (Anderson, 2000, p. 193). The Likert scale is, therefore, more appropriate because apart from psychology, the Guttmann scale has been observed to be more appropriate in political science, anthropology, public opinion and not necessarily social sciences (Anderson, 2000, p. 193). However, its precision and accuracy will still be beneficial in this study.
References
Anderson, L. (2000). Assessing Affective Characteristics in the Schools. London: Routledge
Denzin, N. K. (2006). Sociological Methods: A Sourcebook. New York: Transaction Publishers.
Erwin, P. (2001). Attitudes and Persuasion. New York: Psychology Press.
Frankfort-Nachmias, C. (2007). Research Methods in the Social Sciences (7th ed.). New York: Worth Publishers.
Kumar, R. (2010). Research Methodology. New York: APH Publishing.
Lodico, M. (2006). Methods in Educational Research: From Theory to Practice. London: John Wiley and Sons.
Osgood, C. (1971). The Measurement of Meaning. Illinois: University of Illinois Press.
Plooy, G. (1995). Introduction to Communication. New York: Juta and Company Ltd.
Schoon, I. (2006). Risk and Resilience: Adaptations in Changing Times. Cambridge: Cambridge University Press.
Vaske, J. (1990). Socializing the Human-Computer Environment. New York: Intellect Books.