Class Notes (838,765)
Canada (511,085)
Sociology (204)
SOC 325 (8)
Lecture

March 17(3).docx

4 Pages
127 Views
Unlock Document

Department
Sociology
Course
SOC 325
Professor
Elizabeth Quinlan
Semester
Winter

Description
Soc 325 March 17 2014 1 Questions that more than 40% of the class got wrong on midterm: • Question 24: the central problem in the case of one sample hypothesis test is to determine a. If the sample is random b. If the sample statistics are the same as those of the sampling distribution c. If the parameters are representative of population d. If a sample came from a population with a certain characteristic (found on page 210) • Question 42: based on the output below, what would you say about the difference in the usual hourly wages of employees in the ten Canadian provinces a. There is no difference between the usual hourly wages of employees in the 10 Canadian provinces b. New Brunswickers earn statistically significantly less than employs in other provinces c. Albertans ear statistically significantly more than employees in other provinces d. None of the above (all we can say is there is a difference) • Question 22: in a one-tailed t-test of two sample meals with a alpha of (0.10) what is the z(critical) value a. + 1.64 b. + 1.28 (tableAcolumn C) c. + 0.03 d. + 2.58 Chapter 12: Measures of strength of an association for ordinal variables Gamma, Kendall's tau-b and tau-c for "collapsed" ordinals • Advantage of Gamma - PRE interpretation, downside is it can over exaggerate the measure Logic is based on ordering of pairs of cases (see page 335) • Instead of calculating understand the logic 2 types of ordinal variables --> different measures of association Continuous Ordinal Variables • Spearman's rho, r s is a measure of association of two 'continuous' ordinal variables a. Measures both strength and direction of the relationship b. Ranges from 0 (no association) to ± 1.00 (perfect association) • It won't happen c. Is symmetrical measure • Order of dependent and independent variables does not matter, will still get the same value for interpretation • Logic is slightly different from ordering pairs of cases, similar and dissimilar (pg 360) d. Only when the value of Spearman's rho is squared do we have a PRE interpretation (rs) e. Example: Soc 325 March 17 2014 2 • The number of friends a person is hypothesized to have a positive effect on his/her quality of life • Given the data below, how strong is the relationship between the two variables? Respondent No. of friends Quality of life 1 5 7 2 9 10 3 7 6 4 3 5 5 1 3 6 10 10 7 4 5 8 15 8 9 6 9 10 3 6 11 5 7 6∑ d 2 rs=1− 2 n(n −1) f. The computer value of + 0.84 indicates that there is a very strong and positive relationship between these two variables g. We square 0.84 = 0.71. So, we can say "we make 71% fewer errors in predicting quality of life with knowledge of number of friends" Nominal: measures indicate Collapsed Ordinal: measures indicate strength strength of relation but not and direction of relationship direction Phil: 2 category variables Gamma: has a PRE interpretation Cramers V: variables with any Kendal's Tau-b and Tau-c :has no PRE number of categories interpretation, but does not exaggerate strength of Lambda: has a PRE interpretation as
More Less

Related notes for SOC 325

Log In


OR

Join OneClass

Access over 10 million pages of study
documents for 1.3 million courses.

Sign up

Join to view


OR

By registering, I agree to the Terms and Privacy Policies
Already have an account?
Just a few more details

So we can recommend you notes for your school.

Reset Password

Please enter below the email address you registered with and we will send you a link to reset your password.

Add your courses

Get notes from the top students in your class.


Submit