November 27, 2018 Review Session- Final · 50 multi choice · 30 T/F · Q: A regression analysis is inappropriate when o You have two variables that are measured on either the ordinal or nominal scales · Q: the regression equation for predicting number of speeding tickets (Y) from info about driver age (X) is Y= -. 065(X) +5.57. How many tickets would you predict from a 20 year old? o 4.27 Regression · Q: what does r2 tell us o The proportion of variability in Y accounted for by X . What does it mean to say there is error in our regression? o We cannot predict Y perfectly " It's because Y is always going to leave out some of the dots when we draw a straight line, there's always going to be errors . Does happiness lead to donating more money to charity o DV- donating o IC- Happiness Null Hypothesis- there is no relationship between happiness and donation ( X and Y) o In this case, we can reject the null hypothesis based off the regression table that he showed " The more money you give, the happier you are (positive relationship) " The probability of error is only 1% . We can also find that men give less money than women · Age is not statistically significant · Regression tables IF THERE IS NO ASTIRIX *** NEXT TO IT ON THE REGRESSION TABLE IT IS NOT STATISTICALLY SIGNIFICANT Pay attention to positive or negative relationships !!!! o · Equation for regression table o Y= a + bx+ e o Y = a + 6.83 ( happiness) + e · Q; Janet is studying the relationship between education and income with age held as constant what is the appropriate method of analysis for Janet to use o Partial regression · Q: prof calculate a squared multiple correlation coefficient (aka r2) as being .36. What does this signify o Final score is the DV o 36% of the variance is the final score was explained
November 27, 2018 · R2 is also known as pearson's coefficient or correlation coefficient You just multiply it by 100 and you find out the percentage of the DV that is being explained Field Research · Q: among the advantages of field research Is that it o Allows for modification of research design . Don't have to use a rigid form · Q; suppose we wanted to determine whether municipalities that have no smoking bylaws are more likely to enact laws requiring cyclists to wear helmets than municipalities that do not have no smoking by laws. What is the unit of analysis o The individual municipalities · Q: what is an example of non reactive research o Examine files of students who completed uni 25 years ago · Institutional Research Looking at the ways that people interact with one another · How institutions affect people ( often with marginalized groups) · Particapory Action Research o Working with groups to find problems, and