1 / 27

Measures of Association in Crosstabulations for POLS 300 Study

Understand how to analyze relationships between variables through crosstabulations, using measures like Lambda and Gamma. Learn to interpret these values for insightful research findings.

joelbbrown
Télécharger la présentation

Measures of Association in Crosstabulations for POLS 300 Study

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Crosstab 2 – Measures of Association POLS 300 Butz

  2. Crosstab • Crosstabulations are appropriate for examining relationships between variables that are nominal, ordinal, or dichotomous. • Displays joint distribution of two variables.

  3. Measures of Association • Measures of association summarize efficiently the existence, direction, and strength of a relationship between two variables. • A single summary number.

  4. Measures of Association • Most measures of association range from –1 to +1. • Closer the value to 0, the weaker the relationship. • + and – indicate the direction of the relationship.

  5. Measures of Association for Crosstabulations • Purpose – to determine if nominal/ordinal variables are related in a crosstabulation • At least one nominal variable • Lamda • Chi-Square • Cramer’s V • Two ordinal variables • Tau (tau b, tau c) • Gamma

  6. Lamda • Requires one nominal variable. • Lambda is designed to indicate whether the values of one variable tend “to cluster” with certain values of the other variable • so that knowing a case’s value for the independent variable would help one predict the case’s value for the dependent variable.

  7. Lamda • Lambda= (E without– E with) E without • E withoutstands for the number of errors made in predicting or guessing the dependent variable without knowledge of the independent variable. (Rule 1) • E with is the number of errors with knowledge of the independent variable. (Rule 2)

  8. Lamda • The logic is that if the two variables are related, having knowledge of the independent variable should provide (predict) knowledge of the dependent variable. • Lambda results are a measure of “proportional reduction in error” (PRE).

  9. Lamda

  10. Lamda – Rule 1 (knowledge of dependent variable – partisanship - only)

  11. Lamda – Rule 2(knowledge of independent variable and dependent variable)

  12. Lamda –Calculation of Errors • Errors w/Rule 1: .60 * 100 = 60 • Errors w/Rule 2: 16 + 10 + 14 + 10 = 50 • Lamda =(Errors R1 – Errors R2)/Errors R1 • Lamda = (60-50)/60=10/60=.17

  13. Lamda • PRE measure • Ranges from 0-1 • 0 – means that introducing the IV does NOT reduce errors in predicting DV • > 0 – represents the “percent reduction in error” in predicting DV when IV is introduced!

  14. Relationships between Ordinal Variables • There are several measures of association appropriate for relationships between ordinal variables • Gamma, Tau-b, Tau-c, Somer’s d • All are based on identifying concordant, discordant, and tied pairs of observations

  15. Pairs • Many of the measures for ordinal variables are based on pairing the data. • Concordant pair: one case has higher/lower values on both variables than the other case. • Discordant pair: one case is lower on one of the variables but higher on the other variable. • Tied pair: both observations are tied on at least one of the variables.

  16. Concordant Pairs:Ideology and Voting • Ideology - conserv (1), moderate (2), liberal (3) • Voting - never (1), sometimes (2), often (3) • Consider two hypothetical individuals in the sample with scores • Individual A: Ideology=1, Voting=1 • Individual B: Ideology=2, Voting=2 • Pair A&B are considered a concordant pair because B’s ideology score is greater than A’s score, and B’s partisanship score is greater than A’s score

  17. Concordant Pairs (cont’d) • All of the following are concordant pairs • A(1,1) B(2,2) • A(1,1) B(2,3) • A(1,1) B(3,2) • A(1,2) B(2,3) • A(2,2) B(3,3) • Concordant pairs are consistent with a positive relationship between the IV and the DV (ideology and voting)

  18. Discordant Pairs • All of the following are discordant pairs • A(1,2) B(2,1) • A(1,3) B(2,2) • A(2,2) B(3,1) • A(1,2) B(3,1) • A(3,1) B(1,2) • Discordant pairs are consistent with a negative relationship between the IV and the DV (ideology and voting)

  19. Identifying Concordant Pairs - Obs/cell * Obs. In cells below & to the right - #Concordant = 80(70 +10 +20 + 80) + 20(20 + 80) + 10(10 + 80) + 70(80) == 22,900

  20. Identifying Discordant Pairs • Obs/cell * Obs. In cells above & to the right • #Discordant = 20(10 +10) + 0(10 +10 + 70 +10) + 70(10) + 20(10 + 10) = 1,500

  21. Gamma • Gamma is calculated by identifying all possible pairs of individuals in the sample and determining if they are concordant or discordant • Gamma = (#C - #D) / (#C + #D)

  22. Interpreting Gamma • Gamma = 21400/24400 =.88 • Gamma ranges from -1 to +1 • More concordant pairs will lead to a positive statistic (positive relationship) • More discordant pairs leads to a negative statistic (negative relationship) • Gamma does not account for tied pairs • Tau (b and c) and Somer’s d account for tied pairs in different ways

  23. Identifying Tied Pairs (Row Variable) - Obs/cell * Obs. In cells remaining in row - #Tied1 = 80(10 + 10) + 10(10) + 20(70 + 10) + 70(10) + 0(20 + 80) + 20(80) == 5,600

  24. Identifying Tied Pairs (Column Variable) - Obs/cell * Obs. In cells remaining in column - #Tied2 = 80(20 + 0) + 20(0) + 10(70 + 20) + 70(20) + 10(10 + 80) + 10(80) == 5,600

  25. Tau B and C • Tau b is suitable for square tables, same number of rows and columns. • Tau c is suitable for non-square tables.

  26. Tau B Calculation • 21,400/ sqrt {(24,400 + 5,600)(24,400 + 5,600)} = • 21,400/ 30,000 = .71 • Strong, Positive Relationship

  27. Square tables: Non-Square tables:

More Related