The Traditional Family as an Indicator of Student Success on Arkansas Benchmark Tests

This is my paper that some of you asked to see. I could not figure out how to import the actual word file so you won't be able to see the graphs, charts, etc. If you want to see it (which would be very flattering) just post a comment and I'll Email it to ya.

Got an A in the class by the way.

The Traditional Family as an Indicator of Student Success on Arkansas Benchmark Tests

Background
Many studies have been done to show a relationship between educational outcomes and family structure. Specifically, it is the intention of this study to examine the relationship between students living in a traditional family and the Arkansas Comprehensive Testing, Assessment, and Accountability Program (ACTAAP) scores. Currently, the state of Arkansas requires “Benchmark Exams at grades 3-8, End-of-Course Exams in Algebra I and Geometry, and a Grade 11 Literacy Exam (ADE, 2006).”

The traditional family unit was described by Ginther & Pollak as “families in which all children are the joint children of both parents (2004, 672).” This researcher hypothesizes that students living in traditional families are expected to do better on seventh grade Benchmark Math and Literacy Exams. Furthermore, H0: there will be a statistical significance (α = 0.05) in students living with both their biological mother and biological father. The researcher has noticed from personal work experience that many students in the school district at hand do not live in traditional families. Mere curiosity led to the current study.

Method
Fifty (50) eighth grade students (26 male, 24 female) from an Arkansas delta middle school were given a survey asking their name and who they lived with (Table A1). Most students in the sample are from low income families. More than seventy-five percent of students in the district qualify for free lunch, therefore no students are charged for lunch. Fourteen students (28%) receive some type of special education services ranging from indirect to self contained services. This sample of eighth grade students by no means represents a general population of eighth grade students. There are very few affluent students (if any) in this sample. This will likely skew the results of the study.


Figure A1
Birth Mother/Birth Father
Choices in the survey included birthmother (BM), birthfather (BF), stepmother (SM), stepfather (SF), grandmother (GM), grandfather (GF), aunt (A), uncle (U), brother (B), sister(S), foster-mother (FM), foster-father (FF), adopted-mother (AM), and adopted-father (AF). There was also a space for students to add anyone with whom they live that was not listed on the survey. One student indicated that they lived with their mother’s boyfriend.

Students were insured that surveys would be kept secure but not anonymous. Furthermore, they were told that the only reason for putting their name on the survey would be to connect their survey with their test scores.

After surveys were given, Benchmark test scores were obtained from the previous (2005-06) school year. After scores were put on surveys, names were removed and the list was randomized to insure the students’ identity was removed from the surveys and test scores.

Results
All data analyses were performed using SPSS version 14.0 (SPSS 2005).

When surveys were tallied, the results were not promising. A larger, more diverse sample was clearly needed. As one with average intelligence can gather the range and standard deviation was all over the place for Benchmark Math & Literacy scores (Table A2).

Chi-Square test (Table A3) revealed that it is highly unlikely that in a normal population, the two groups, students living with both biological parents and students not living with both biological parents are not equally as likely.

It was my hope that this study could use a level of significance of 0.05. However, after running independent sample t-tests on SPSS software (Table A4), we would have to use p < 0.861. This is in no way ideal. The outcome of this study can be highly credited to sampling error. There are many other variables that were not or could not be tested. For example, SES data could not be collected because of liability issues with the school and students (even if they could) would not give out that kind of information with any reliability.

Pearson Correlation Coefficients (Table A5) were not significant in math or literacy when students living in traditional families were compared to students living in traditional families (r = 0.071 & r = -0.025 respectively). This suggests that living in a traditional family has little impact on Benchmark test scores.

Despite the hypothesis being rejected, the researcher still strongly holds to the belief that students living in traditional families do better on the Benchmark test for several reasons:
1. The sample was too small
2. The sample was taken from only one school district in the Delta Region of Eastern Arkansas
3. Many other variables were present that were not able to be tested.

Future studies in this area would benefit from including a more diverse selection of geographical regions of the state. This would be beneficial because it would include a wider range of students. A diversity of race, SES, and culture would give a more accurate picture of the problem. A researcher would need to have access to multiple resources including state databases; current census data; and many willing, honest participants.

Another limitation of the research is the fact that the Benchmark scores at the school are generally low (Math μM = 586.30 and μL 534.12). Future research would benefit from not only looking at Benchmark Scores but daily grades as well.

While no definitive results were found, this area is a wide open arena for the state of Arkansas. A more expansive study would be very valuable for educators. From that study, interventions for math and literacy could be derived and implemented for the benefit of the students of Arkansas.