Research Survey Design, Coursework Example
The research survey will focus on exploring the relationship between a student’s grades (as a proxy of performance) and various factors that influence a student’s performance in school. This is a topic of increasing interest as federal and state policymakers not only consider policy changes in cutting the school day/ year, but also cutting after-school programs, extracurricular activities, and putting more onus on the parents to be active in the student’s educational process. While the survey will not be able to address claims of how a shortened time in school will affect a student’s academic performance, it will present findings that can inform debate on what factors may be predictive of a student’s performance, and thus which cuts may be more harmful than others .
The population for the survey will include boys and girls from the ages of 10-22. The survey includes three types of students (middle school, high school, and college)- there is a wide range of students included in the survey so that their performance can be traced over a significant time period. The survey will implement a longitudinal design in order to understand what factors are impacting a student’s performance at a particular time starting in middle school, and how those factors interact with academic performance from middle school through college. Although it would be optimal to begin surveying students in elementary school (or even earlier) in order to understand how early factors impact performance, the funding and logistics for such a study are outside the scope of this survey.
The sampling method will compose of a simple random sample students located in four different regions throughout the United States east, west, south, north. The individuals will be selected based on state records of enrolled high school students in the selected states. In total, 5,000 students will be selected in each region, 2,500 boys and 2,500 girls to ultimately participate in the study. In order to get the necessary 5,000 participants in each region, more students will need to be selected initially, with boys being oversampled as they traditionally have a lower participation rate than females in school surveys. Once the students are selected, a letter will be sent to their parents requesting participate in the study over the course of 12 years: four years in middle school, four years in high school, and four years in college.
The survey’s main dependent variable will be the student’s academic performance (the student’s gpa). The student’s grades will be taken directly from official transcripts at the respective institution. The survey will then focus on assessing and conceptualizing four main independent variables in predicting a student’s academic performance: 1) parental involvement; 2) use of spare time; 3) eating habits; 4) exercise habits. The “parental involvement” variable will focus on how much time the parent spends with the child doing homework, reading to them, and their overall concern with their student’s progress (taken from the parent survey results). The variable “use of spare time” will assess which extra circular students the student is involved in (including sports, clubs, and music). The variable “eating habits” will consist of a nutritional survey that assesses what the student usually eats and at what times- the survey will also assess if the student receives federal assistance in school lunches or not. Finally, the variable “exercise habits” will assess how much exercise a student gets and at what intervals at a weekly basis. Because there is no one unified instrument that addresses all the independent variables, it will be developed along with existing scales in the area of grade and parental influence.
Once the students are selected to participate in the survey, the students will be given the survey annually over a two-to three week period. In order to validate the student’s survey vis-à-vis their habits, the parents will also be surveyed to understand the students’ answers. Overall,
Once the surveys are completed, data analysis will be completed on an annual basis to assess the student answers. Although the initial analysis will focus on gaining an aggregate picture of students across the education spectrum, the data will also be stratified in order to assess what factors play a key role in the development of boys and girls at different times in their educational careers. Thus, a descriptive analysis will be done first to understand which of the predictors proved to be significant in predicting academic performance; once that is completed, a more detailed analysis will be given looking at the stratified data. In addition, the questions for each variable will ultimately be collapsed into an aggregate scale for each of the four dependent variables. There will also be statistical tests conducted to check the validity of the self-reported data, as well as studies that try to focus on the validity of the scales used to capture certain variables. The data will finally be analyzed via linear regression that attempts to predict which factors can influence a student’s academic performance from middle school through college.
Due to the size and scope of the survey, there will likely be a number of threats to validity that will need to be watched for. In particular, there will be questions of selection and response bias; that is, does the survey really capture a broad enough cross-section of US students that is representative of students in general. While the large sample size (n=20,000) and the oversampling of males is meant to mitigate these threats, there is still the probability of having some bias from students who may not be representative- such as having too many suburban versus urban students. The results will be analyzed annually in a paper submitted to a prominent educational journal, as well as a presentation given to inform educators of the survey’s findings.
Cresswell, J. (2003). Research Design: Qualitative, Quantitative and Mixed Method Approaches. Thousand Oaks, CA: Sage Press.
Fink, A. (2008). How to Conduct Surveys: A Step-by-Step Guide. Thousand Oaks, CA: Sage Press.
Kin, R. (2008). Case Study Design: Research and Methods. Thousand Oaks, CA: Sage Press.
Johnson, B. & Christensen. (2010). Educational Research: Quantitative, Qualitative, and Mixed Approaches. Thousand Oaks, CA: Sage Press.
 It is impossible to know what the total number would be; this would depend on the power needed for the study.
Time is precious
don’t waste it!