Class Notes (836,580)
Canada (509,856)
Sociology (4,081)
SOC222H5 (93)
Lecture

SOC222 Lecture 2

12 Pages
123 Views
Unlock Document

Department
Sociology
Course
SOC222H5
Professor
John Kervin
Semester
Fall

Description
SOC 222 -- MEASURING the SOCIAL WORLD TABLES -- Session #2 Sep 16/2013 Topics: Review Today’s Objectives Frequency Distributions Frequency Distributions in SPSS Crosstabulations of Two Variables Percentaging Tables Elaboration: Adding a Third Variable Graphs: Bar Charts Bar Charts for Frequency Distributions Bar Charts for Crosstabs Readings: Linneman, ch. 2 1. FREQUENCY DISTRIBUTIONS Frequency distributions tell us the number of cases • In each category • Of a category variable Category variables are ordinal or nominal numbers. Level of measurement – drives statistical test you can use Why would we look at just one variable? 1. Own variation in the variable – does it change or is it the same for everyone? A variable that varies can co-vary with another variable, which means we can check if they correlate or have a cause-and-effect relationship. 2. Distribution – what a variable looks like; goes beyond sociology  look at voting, test grade distributions, percentage of female law school applicants Here’s the distribution of marks for test #1: SOC 222: Frequency Distribution of Marks for Test #1 Letter Grade Frequency Percentage A 54 40.0 B 42 31.1 C 31 23.0 D 6 4.4 F 2 1.5 Total 135 100.0 Source: BlackBoard The table shows: Frequency = count, which is percentage in the column next to it Frequency distribution There is enough distribution for the marks in test 1 to be used as a variable because not even close to everyone was in the same category, so it is not the case that everyone is getting the same grade Frequency Distributions in SPSS New data set: “Students-1” New document on web site: SPSS Frequency Distributions Questions: • The students were surveyed in different departments. Which departments were students in? We want to know how many departments there are and how many students came from each one. • How many in each dept.? Enough variation to use as a variable? In SPSS: “Analyze” dropdown menu “Descriptive statistics” “Frequencies” This opens a box called “Frequencies” • On the left: List of variable labels • Variable names (depart) VS variable labels (Department) • In the lists, the variable name is given in [ ] square brackets! • Below the variable list: a check box: IGNORE • In the middle: an empty working area called “Variable(s)” • Option buttons on the right: IGNORE • Action buttons on the bottom • OK to make the table • RESET to start over again TRICK: Expand the box sideways to read the variable labels Click on the variable you want Click on the arrow to move it to the Variables area (OR drag and drop) Click on OK This opens your output window, and there’s your frequency distribution table. Department Frequency Percent Valid Percent Cumulative Percent Valid Chem Eng 29 10.2 10.2 10.2 Elec and Comp Eng 47 16.5 16.5 26.8 East Asian Studies 39 13.7 13.7 40.5 English 20 7.0 7.0 47.5 Microbiology 63 22.2 22.2 69.7 Botany 19 6.7 6.7 76.4 Geography 11 3.9 3.9 80.3 Economics 38 13.4 13.4 93.7 Sociology 18 6.3 6.3 100.0 Total 284 100.0 100.0 Notice: The frequency distribution table: • Shows the count, or frequency, in each category of your category variable • IE, how many students from each department in the sample of 284 • The title is the variable label • It gives totals at the bottom • It shows the percent of the total • It shows the valid percent • This percentage omits any missing data • Not important at this point • And this variable has no missing answers • It shows the cumulative percent (adds up valid percent) • Only useful for ordinal category variables • See Linneman text, p. 40 What this frequency table tells you: Count, percentages Tells you survey covered 9 departments Geography had the least amount of students, microbio had the most There is enough variation; if we had the same # of students we would have the most amount of variation (max) Readings: Linneman: 39-43 Note: Kranzler ch. 4 has same title • But it’s about ratio variables • Ignore it for now 2. CROSSTABULATIONS of TWO VARIABLES • New doc posted on website: “SPSS Crosstabs” RQ: Do students with high grades like being at UofT more than other students? • IE: do high marks affect your opinion of UofT In SPSS: “Analyze” dropdown menu “Descriptive statistics” “Crosstabs” This opens a box called “Crosstabs” • On the left: List of variables • Below: two check boxes: IGNORE • In the middle: three empty working areas called “Rows”, “Columns”, and “Layer” • IGNORE the “layer” area for now • Option buttons on the right • You’ll use “Cells”, IGNORE OTHERS • Action buttons on the bottom • OK to make the table • RESET to start over again Decide on your DV • For our RQ: it’s “Like being at UofT” • Survey item is: “I am glad I chose UofT – agree, unsure, disagree” • Variable label is “Glad I chose UofT” • Move it to the “Rows” area Move your IV to the “Columns” area • Students with high grades • Survey item is “My average last year” • This has been recoded into two categories – “B+ or better” and “other” • Variable label is “B+ or better” • Move it to the “Columns” area Click on OK This brings up your crosstab table Glad I chose UofT * B+ or better Crosstabulation Count B+ or better Lower than B+ B+ or better Total Glad I chose UofT Disagree 36 28 64 Unsure 41 32 73 Agree 56 76 132 Total 133 136 269 Is there a relationship? • Hard to tell • No numbers that tell us if there’s a relationship Just looking at this, we can’t say Percentaging Tables • The rules: 1. DV categories go in rows 2. IV categories go in columns 3. Percentage down -- within columns (within IV categories) 4. Compare across – within rows (within DV categories) Readings: See Linneman pp. 46-47 In SPSS: In the Crosstabs box • After you’ve moved your DV into “Rows” area, and IV into “Columns”… • Click “Cells” button on right • This opens a sub-box called “Crosstabs: Cell Display” • This lets you choose what goes into each cell of your table • Counts of #s of cases automatically go in • You want to add percentages • Sub-box has with five separate areas • IGNORE ALL except “Percentages” • The “Percentages” area gives you three choices 1. Row 2. Column 3. Total • We want to percentage down • IE percentages within each column • So we select “Column” • This gives us percentages within each category of the IV (marks) • We can then compare across • Within each category of the DV (feelings about UofT) • Read Linneman, pp. 46-47 • Click “Continue” to close the Cells sub-box • Click OK • This brings up the same crosstab table • Except this time with percentages Glad I chose UofT * B+ or better Crosstabulation B+ or better Lower than B+ B+ or better Total Glad I chose UofT Disagree Count 36 28 64 % within B+ or better 27.1% 20.6% 23.8% Unsure Count 41 32 73 % within B+ or better 30.8% 23.5% 27.1% Agree Count 56 76 132 % within B+ or better 42.1% 55.9% 49.1% Total Count 133 136 269 % within B+ or better 100.0% 100.0% 100.0% What does this crosstab table tell us? • TRICK: • Choose the one most interesting or important (the one that makes the most sense) category row in the DV (Glad) • This would be the “Agree” category • Those glad th
More Less

Related notes for SOC222H5

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