Use the set of 17 questions that were asked of child’s teacher at the end of kindergarten about the child’s skills. Use these to run factor analysis and reliability in order to create one or more scales. Each individual variable is a teacher’s rating in the areas of literacy, or math. Teachers rated students as 1=Not yet, 2=Beginning, 3=In Progress, 4=Intermediate, 5=Proficient. Students who were rated as 6=non-applicable were recoded to have a score of 1. These are ordinal variables that we will treat as continuous.
a) Examine how the items in the questionnaire data set (attached) were asked. Then examine these variables in SPSS. Take a look in variable view. Run the appropriate descriptive statistics (frequenices/percents, or mean and SD) on them. Note: These variables have a T2 at the beginning indicating they are teacher report from wave 2 of data collection (spring of kindergarten). Make sure you use this version NOT
First, conduct an open factor analysis using principal components analysis and a varimax rotation. Paste the output below. How many factors did SPSS extract? How do you know how many factors were extracted?
Next, paste your rotated component matrix here or make your own table. Highlight or use some other method to show which variables load on to each factor. Do you think they make sense? What would you call each factor?
Next, force a 1 factor solution. What are the results? Do all variables load “high” enough on to one factor? How do you know? Paste the output and explain below.
Now measure internal reliability for the new scale if you included all 17 variables together. What is the chronbach’s alpha? Is it low, moderate or high? What does it tell us about this new scale?
Create the new scale as a mean score by creating a new variable that adds up all of the scores and divides them by 17. Make sure you name your variable in a way that makes sense that you could remember. What did you name your variable? Run descriptive statistics on this variable. What is the mean and standard deviation. Create a histogram. Is the variable normally distributed or skewed?
Now that you have a new scale that is a continuous measure of teacher reported skills at the end of kindergarten, use this variable to run a regression analysis examining whether the number of books a child has in their home is predictive of teacher reported academic skills at the end of kindergarten (your new variable).
Dependent Variable: new variable you created (continuous variable)
What type of regression should you use and why? Paste the relevant output tables below.
What are the results? What can you say about the model and the relationship between the independent and dependent variable?
a) Find the validity and /or reliability information in the CASAS assessment Test and TABE testing based on research studies using it or a specific report or study to “validate” the instrument. What is the instrument? how it is administered, what construct(s) it measures, who the reporter is and anything else that would be helpful to know.
b) What information did you find about reliability and validity of the instrument? Where did you find it?
c) Would you recommend that this instrument should be used in future research? Why or why not?
d) Pose a question about the instrument or instrumentation in general.