Statistics Discussion Response/Social Science

Statistics Discussion Response/Social Science

Textbooks

1. OpenStax, Introductory Statistics. OpenStax. 19 September 2013. http://cnx.org/content/col11562/latest/

2. Tyrrell, S. 2009. SPSS: Stats practically short and simple (1st edition). bookboon.com. Ebooks and textbooks from Bookboon.com https://oerstatistics.wordpress.com/2016/03/05/spss-books/

Original Question:

Provide a brief introduction to your study to remind your classmates what we are reading about here.  You will also title your thread accordingly (do not include the week # or your name).  This week we talk about the uses of a crosstabulation (crosstab) and the benefits of creating this “snapshot” of your data.  Create a crosstab for your data and include in the post.  Be sure to explain your findings.

You will also identify the following times about your study:

1. Your overall research question;     2. The research hypothesis and null hypothesis

Special note:

When a variable is continuous (interval/ratio level of measurement), for example, age of respondent, we do not run crosstab directly b/c it will result in a really spread-out table with lot of 0s and low frequency cells. Such crosstab does not help us understand the data.

The correct way is to reduce the level of measurement to either ordinal level or nominal level and then run the cross table. The way to fix is to reduce I/R level of measurement to a lower level by lumping columns in into just a few categories. For example, you can group age 19 to 22 as one category, 23 to 30 as one category, and so on.  In this way, your crosstab will help us better understand data. Here is an example of recoding: https://www.youtube.com/watch?v=uzQ_522F2SM

Interpreting crosstab output

When you interpret the result, please include the discussion of epsilons (p. 170) and the 10% rule (p. 165). The epsilons in short is the differences between the highest and lowest column % in any given row.  As long as one epsilon makes the 10% threshold, we’ll deem two variables have “enough” going on to with each other to warrant further statistical analysis.

Reply to the following response with 200 words minimum. (please make response as if having a conversation, respond directly to some of the statements in below post.)

For my research study I am trying to determine whether or not the number of children a people have will affect their opinion about education spending.

The research question for my study is “What affects people’s opinions on education spending?”

Variables:

1. “nateduc” which stands for improving nations education system (DV)

1. “childs” Number of children? (IV)

Research Hypothesis:

H1: People’s opinion on education spending is determined by the number of children they have.

H0: People’s opinion on education spending is not determined by the number of children they have.

Crosstab Interpretation

Based on the data there were 733 (74.5%) individuals who believe there is “too little” money spent on education, 173 (17.6%) who believe that spending on education “is about right”, and 78 (7.9%) of people who believe there is “too much” spending on education. Cross tab analysis revealed that approximately 78.7% (111/141) of the respondents with one child believe there is “too little” spending on education. While 76.3% (106/139) individuals with 3 children and 76.2% (208/273) of individuals with 2 children believe there is “too little” spending on education. However, 74.6% (226/303) individuals with no children also believed that there is “too little” spending on education.

The epsilons for the number of people who said that there is “too little” money spent on education is 62% (78.7% -16.7%). While the epsilon for the number of people who said spending on education is “about right” was 59.3% (66.7% -12.8%). The epsilon for individuals who said it was too much spending on education was 16.7% (16.7%-0.0%). Based on this information the dependent variable “nateduc” has a strong effect on the independent variable “childs;” therefore based on the 10% rule this relation is worth further examination.

EXAMPLE OF ANOTHER REPLY:

hat’s interesting to me about your cross tab analysis is that families with 0 thru 3 children are generally consistent in their responses to the question of education spending.  In other words, this grouping of respondents overwhelmingly believes that there is too little spending on education.  However, and interesting slope change occurs with families with 4 to 6 children.  If I’m to trust the veracity of the GSS data set, then these families are less convinced about the spending issue…albeit, not that dramatic of a drop.

What’s most shocking though is that families with 7 or more show a severe drop in their belief about spending on education.  This grouping of families are seemingly satisfied (about right or too much) with spending on education.

How can this be?  Why do “smaller” families see chronic underspending where large families do not?  Further, large families (7 or more) have the most to gain with spending…and if anything, are a contributor to the “problem”…they have more kids to educate that everyone is collectively paying for.  Said another way, their benefit in terms of dollars spent per child is much higher than smaller families.  Or, is it a question of their cultural priorities…that larger families’ parents are less educated and therefore see less value in education than smaller families?  Conversely, smaller families’ parents are more educated and see a greater value in obtaining a quality education? So many questions I have!  I’m not sure where your research will take us or conclusions you may draw, but I’m sure they will be interesting!