LIR 832 SPRING 2007 computer problem set 2

LIR 832 SPRING 2007 computer problem set 2 - Computer...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
Computer Problem Set #2 Spring 2007 Tables, Covariances and Correlations We will be using the 20% sub-sample of the Manager and Professional data set that you used in your last computer problem set. You will also want to refer to the UNICON codebook as this provides definitions of the variables we will be using in this problem set. 1. Creating Tables. Minitab has the capacity to create complex tables and summarize data within the categories in the tables. Consider the distribution of managers and professionals by marital status and by type of employer. a. Marital status is coded as 1 married 2 married (spouse absent) 3 widowed 4 divorced 5 separated 6 single never married Recode this data into married (combine (widowed, divorced, separated) and single never married. (Call this never married) Code these as text variables ( Data < code, numeric to text ) Title this variable marital status MarStat. You will need to associate each number with a text description even if it is repetitive. b. Next we are interested in recoding workcls1 into private, federal, state, local employer or self employed using numeric to text ( Data code, numeric to text ). The coding for this variable is 1 Federal worker 2 State worker 3 Local worker 4 Private for profit 5 Private non-profit 6 Self Employed 7 Self Employed 8 Without pay Before doing this, use the tally command (Statistics Table Tally) to get the number of observations for each of the outcomes. Note that we are a group, “without pay” that is very small and that we likely don’t care much about. We would like to remove this group from our data and to do that, we are going to do a two stage recode; first converting workcls1 variable so that the “without pay” code is missing data ( ‘*’), and then taking the new variable and recoding it into
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
text data for type of employer. First, using the code command ( Data < code, numeric to numeric ) we will recode from workcls1 to temp. Do a numeric to numeric recode with “8" as the original value and “*” – missing data, as the new value. Note that no other values of workclas1 will change. Next do a numeric to text recode placing appropriate names in for the federal, state, local, private and self-employed values. Don’t do anything with the missing data, leave that blank. By the way, you can place 4:5 in the original value position for private and both will be coded as private. Similarly for 6:7 for self-employed. c. Now create a table with the number of observations in each of these categories. To do this, go to Stat Tables cross-tabs and chi-square . Make the classification variables so that EmpType are the rows and MarStat are
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

Page1 / 6

LIR 832 SPRING 2007 computer problem set 2 - Computer...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online