exam 2k17 - 2

Question 1

experimental vs observational

Answer 1

observational = measured but did not intervene

• broader aim of research was experimental as using the app to try increase talk time, but this was not the focus of this study

Question 2

scatterplot

- association

Answer 2

look at if neg or positibr,
strength (not, mod. or very) and if linear (how much eg very, roughly or not

line is the predicted values and the dots are the observed values
- the higher the R-squared, the better the model fits your data
w/ 0 = no variability of response data around mean
and 1 = mean models explains all variability of the response data around the mean

Question 3

SPSS output of linear regression analysis for relationship btwn … and …. What is teh correlation coeffecicient for this relationship?

Answer 3

R values

- generally under model summary

Question 4

what does r square value tell

Answer 4

% variability in ….. what you’re hypothesis is/what trying to show difference

Question 5

what is the estimated increase in … what you’re showing… associate w; a one unito increase in … what are the units f the estimate ?

Answer 5

go to coefficientsa table > B colun and then the row you think you want to fin is

Question 6

normal p-plot of a regression standardized residual

• what is a residual

Answer 6

the diff. btwn the observed value (dot) and the value predicted by the model (linear line)

for a set number, to find out an estimated residual, see difference btwn point away from line

Question 7

normal p-plot of a regression standardized residual

- regression assumption is plot checking

Answer 7

whether the plot checks the normality assumption

- no strong pattern away from line so normality seems reasonable

Question 8

normal p-plot of a regression standardized redisuals

null and assumptions

Answer 8

null = the slope of the association between the… and ..( two variables) is 0

can find p values = at coefficients a table - sig. colum and variable row