Tori Gagliardi’s E-portfolio

Over the course of this semester, I have taken the class Stats in the Social Sciences which is a beginner course in statistics in the social sciences. We have learned a variety of statistical tests that helped us analyze, interpret, and understand the significance of our data. Not only did we learn how to conduct these tests by coding them into our computers but also how to perform them by hand. This course has expanded my knowledge greatly in terms of statistics in the social sciences and has taught me how statistics come into play in order to read and understand the data.

In Stats in the Social Sciences, we were taught how to use the statistical systems, R-Studio and SPSS. While using these systems we learned how to code and recode variables in order to obtain the data that we desired. In order to use R-Studio, we had to enter codes in order to receive our information. These codes made it possible for us to run tests such as ANOVA, T-tests, Regression, Correlation, finding mean and standard deviation, and Chi-squared. We were then taught how to read, analyze, and understand the data that these tests produced. This required critical thinking because the data can be difficult to read. When using SPSS, we learned how to input information into the system, code and recode variables, and interpret graphs and other data that was shown. For several of these tests, we were tasked with determining the significance, or lack thereof, of the data. In order to do this we had to analyze the data and be able to tell if the data was or was not significant at the .05, .01, and .001 levels. The significance of the data helps to answer our research questions by confirming or denying if it is in fact significant data. Both of these statistical systems will be useful to have in my back pocket if I am conducting other research because it is a quicker way of obtaining the data that you are seeking and can help you better understand your data. 

We were not only taught how to use these statistical systems in order to conduct tests but we were also taught how to conduct these tests by hand. We were shown how to do independent and dependent sample t-tests by hand which requires plugging data into formulas and doing math in order to compare the t-ratio with the critical value. We were able to do this by determining the degrees of freedom and the level of significance which again helped us to better understand the data. In order to better understand the significance of our data we also learned about p-values and how to compare them to the levels of significance and determine if the data is or is not significant. We were also taught to interpret the null and research hypothesis and how to analyze them to understand if we retain either hypothesis or reject it which again helped to interpret the data we produced through the test. We were also taught how to conduct one-way ANOVA tests by hand. This test produced the sum of squares and the f-ratio, both of which measure variation in some form. Through the steps that we learned we were then able to read and interpret what our data meant and again be able to retain or reject the null or research hypothesis and determine the significance between means. 

Two more tests that we were taught by hand are the one-way and two-way chi-squared tests and the correlation test. For the chi-squared tests, we were tasked with doing math, setting up a summary table, finding the degrees of freedom, comparing the chi-squared value to the critical value table, and again determining if we should reject or retain our null or research hypothesis and the significance of the data. Lastly, we were taught two formulas in order to determine if there was a strong or weak and positive or negative correlation. In order to do this we were tasked with finding the r statistic which helped us determine the correlation of the data. Using another formula we were able to determine if the data was or was not significant and if we retain or reject the null or research hypothesis. 

In the future, if I am conducting research again I will be able to pull from the information  that I have learned in this class in order to interpret and analyze the data that I would have. Being  able to use R-Studio and SPSS is an advantage that I will now have that not all students have when graduating college. A future employer may potentially see that as a benefit to hiring. In order to best understand your data being able to determine the significance of it is extremely important and I have been taught how to do so in several different ways. All of these tests and both statistical systems are ways in which you can help answer any research question that includes an independent and dependent variable and better understand your data and potentially find answers that you are looking for.

Throughout the course of the semester, I have been performing tests in order to answer the research question, “How does household income affect family involvement?” This question comes from research conducted with The Andy Taylor Center and Head Start. These are both child care centers whom we sent activities and surveys in order to better understand family involvement. Three of the tests that were run in order to answer the research question are the ANOVA test, the Regression test and the Chi-squared test. These were conducted using both the R-Studio and SPSS statistical systems.

The independent variable question for the research conducted is, “How involved was your family throughout the activity?” This question could be answered by participants on a scale of 0 to 10 with 0 meaning not at all and 10 meaning a great amount. This question was recoded for some tests due to their being multiple questions involving parent involvement levels and the necessity in order to run the test.

The dependent variable question for the research that was conducted is, “What is your annual household income?” This question could be answered with the categories, “Less than $10,000”, “$31,000 to $50,999”, “$51,000 to $70,999”, “$71,000 to $90,999”, “$91,000 or more” and “Prefer not to answer.” The option of “Prefer not to answer” was missing data and was recoded to not be included in the tests that were conducted.

Table 1.

Chi-squared

	Low Income	High Income
Not Engaged	4	4
Moderate Engagement	16	4
Engaged	12	12
X-squared= 4.68	df=2	p-value= .09633

This table shows that in the “Not Engaged” category there were four respondents that were in the low income bracket and four respondents that were in the high income bracket. The  “Moderate Engagement” category shows that there were 16 respondents in the low income bracket and four respondents in the high income bracket. In the “Engaged” category there were 12 respondents in the low income bracket and 12 respondents in the high income bracket. We also see that the chi-squared statistic is 4.68. The degrees of freedom for this test is 2. The p-value shows us that there is no significant difference at the .05 level. This test was recoded to only include low and high income levels removing the six different categories of answer choices. This test was conducted using R-studio.

Table 2.

Basic Linear Regression

	Estimate	Std. Error	t-value	Pr
Intercept	7.2241	0.7167	10.079	2.88e-15***
v36	-0.1034	0.1603	-0.645	0.521
R-squared= 0.005917

This table shows a Basic Linear regression in R-studio. V36 is standing in place for the dependent variable, “What is your annual household income?” Family involvement has a negative correlation with household income. For every one unit family involvement increases, household income decreases by -0.1034. There is a significant finding at the .001 level. The R-squared statistic is 0.005917 which means that income has explained .59% of the variation in family involvement. This test was conducted using R-studio.

Table 3.

ANOVA

	Sum of squares	df	Mean Square	F	Sig.
Between groups	255.259	6	42.543	3.773	.003
Within groups	687.800	61	11.275
Total	943.059	67

This is an ANOVA test run in the statistical system SPSS. Family involvement was measured using the question, “How involved was your family throughout the activity?” We can see that the significant value is .003 which below the .01 level and therefore is significant at the .01 level. This means there is a significant difference between the means.

In order to answer the research question, “How does household income affect family involvement?”, these three tests were conducted. The independent variable being family involvement and the dependent variable being household income. In the Chi-squared test we see that there is no significant difference at the .05 level between family involvement and household income. With the Regression test, it shows that family involvement has a negative correlation with household income. Finally, the ANOVA test shows that there is significance at the .01 level for family involvement. Overall, we see that household income does not have a significant effect on household income.

Welcome to Longwood Blogs. This is your first post. Edit or delete it, then start blogging!