preview

STEM Careers

Decent Essays

Another variable I looked at was Worthiness. I would have preferred to measure self-esteem, but again the dataset did not include this as a variable. The variable worthiness was labeled as I am a person of worth. Table 3 shows that it was operationalized from 1 to 5 ranked from strongly disagree to strongly agree. This is an ordinal variable with 5 categories that can be treated as an interval ratio variable. The measure of central tendency used would be the mean and standard deviation. As seen in Table 3, the mean is 4, meaning that the average response was 4) Agree and the standard deviation is .78, meaning that 68% of the responses were in between 3.22 and 4.78, because the standard deviation was high the responses were more scattered from …show more content…

I did this by looking at the variable 2011 STEM occupation, its label was STEM Career 4 Categories 2011. I recoded this variable into Non-STEM and STEM careers, making this variable nominal. After the recode this variable was coded as 0) being Non-STEM, and 1) being STEM careers. As seen in Table 4, the measure of central tendency used was mode, which was 0, meaning that the majority of respondents chose 0) Non-STEM. The variable 2011 STEM occupations, has a high reliability and validity, this is true because the variable was found in the same codebook, and it measures whether the student had obtained a career in STEM by …show more content…

Because this was a longitudinal study there were a lot of repetitive questions, the same question asked each year for the study, but the responses were not organized in a manner that made it easy to decipher. The codebook lacked the survey questions, which made identifying what was actually asked to the original participant rather difficult. This meant that it was necessary to infer the conceptualization of each variable, which brought down the reliability of the whole study. The dataset was set up alphabetically, and the codebook provided a guide for using it, each letter represented a year, the second letter represented whom the respondent was (12th grader, teacher, principal, etc.), which was straightforward, but at the end of the dataset they added only a few of the original questions, throwing off any and all organization. Another limitation was that the Asian population in this sample was very small; this makes it hard to generalize any of the findings to the general public. This is problematic, because I wanted to focus on Asians, yet without a large sample size it is hard to get a representative sample. The next limitation is the fact I could not

Get Access