1 / 41

# Elaboration of Bivariate Tables: A Direct Relationship - PowerPoint PPT Presentation

Elaboration of Bivariate Tables: A Direct Relationship. Creating the Crosstabulated Table - 1. In SPSS, you can obtain a crosstabulated table by clicking: Analyze > Descriptive Statistics > Crosstabs

Related searches for Elaboration of Bivariate Tables: A Direct Relationship

I am the owner, or an agent authorized to act on behalf of the owner, of the copyrighted work described.

## PowerPoint Slideshow about 'Elaboration of Bivariate Tables: A Direct Relationship' - Pat_Xavi

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.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.

- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript
Elaboration of Bivariate Tables:A Direct Relationship

In SPSS, you can obtain a crosstabulated table by clicking:

Analyze > Descriptive Statistics > Crosstabs

In other words, first, click on Analyze in the menu bar of SPSS, then find and click on Descriptive Statistics on a pop-down submenu. Again, find and click on Crosstabs on another pop-down submenu. Then, you will see a window shown in the next slide.

Move the variable [owngun] to the “Rows(s):” list box.

Move the variable [conserv] to the “Columns(s):” list box.

Click on “Cells…” button on the bottom.

Mark the Column check box in the Percentages panel.

Click on the Continue button to close the dialog box.

Click on the OK button to produce the output.

There is a relationship in the bivariate table if the absolute value of the difference in column percentages on one of the rows is 10% or greater.

In this example, the difference is 57.6% - 73.1% = -15.5%. There is a relationship.

To add the control variable to the analysis, click on the Dialog Recall tool button and select the Crosstabs command from the pop-up menu.

To add the control variable to the analysis, move the [sex] variable to the list box in the Layer panel.

Click on the OK button to produce the output.

The partial table for the first value or condition of the control variable (sex = male) is in the top half of the three-way crosstabs table.

The partial table for the second value or condition of the control variable (sex = female) is in the bottom half of the three-way crosstabs table.

There is a relationship in the first partial table if the absolute value of the difference in column percentages is 10% or greater for the same row we compared in the bivariate table.

In this example, the difference is 55.2% - 71.2% = -16.0%. There is a relationship in the first partial table.

There is a relationship in the second partial table if the absolute value of the difference in column percentages is 10% or greater for the same row we compared in the bivariate table.

In this example, the difference is 60.0% - 74.6% = -14.6%. There is a relationship in the second partial table.

To compare column percentages, we compute the difference between the lower left-hand cell in the partial tables and the bivariate table

The percentage in both the first partial table (71.2%) and the second partial table (74.6%) are LESS than 5% different from the bivariate table (73.1%).

To compare column percentages, we compute the difference between the lower right-hand cell in the partial tables and the bivariate table

The percentage in both the first partial table (55.2%) and the second partial table (60.0%) are LESS than 5% different from the bivariate table (57.6%).

Elaboration of Bivariate Tables:A Direct Relationship

With the information about the relationships in the partial tables and differences in column percentages, we can answer the question.

Elaboration of Bivariate Tables:Detecting a Spurious Relationship

In SPSS, you can obtain a crosstabulated table by clicking:

Analyze > Descriptive Statistics > Crosstabs

In other words, first, click on Analyze in the menu bar of SPSS, then find and click on Descriptive Statistics on a pop-down submenu. Again, find and click on Crosstabs on another pop-down submenu. Then, you will see a window shown in the next slide.

Move the variable [prayer] to the “Rows(s):” list box.

Move the variable [minorty] to the “Columns(s):” list box.

Click on “Cells…” button on the bottom.

Mark the Column check box in the Percentages panel.

Click on the Continue button to close the dialog box.

Click on the OK button to produce the output.

There is a relationship in the bivariate table if the absolute value of the difference in column percentages on one of the rows is 10% or greater.

In this example, the difference is 24.0% - 38.5% = -14.5%. There is a relationship.

To add the control variable to the analysis, click on the Dialog Recall tool button and select the Crosstabs command from the pop-up menu.

To add the control variable to the analysis, move the [fundamen] variable to the list box in the Layer panel.

Click on the OK button to produce the output.

The partial table for the first value or condition of the control variable (religious orientation = not fundamentalist) is in the top half of the three-way crosstabs table.

The partial table for the second value or condition of the control variable (religious orientation = fundamentalist) is in the bottom half of the three-way crosstabs table.

There is not a relationship in the first partial table since the difference of 42.9% - 41.0% = 1.9% is less than 10%.

There is not a relationship in the second partial table since the difference of 21.4% - 25.9% = -4.5% is less than 10%.

Elaboration of Bivariate Tables:Detecting a Spurious Relationship

With the information about the relationships in the partial tables and the bivariate table, we can answer the question.

Elaboration of Bivariate Tables:Detecting a Conditional Relationship

In SPSS, you can obtain a crosstabulated table by clicking:

Analyze > Descriptive Statistics > Crosstabs

In other words, first, click on Analyze in the menu bar of SPSS, then find and click on Descriptive Statistics on a pop-down submenu. Again, find and click on Crosstabs on another pop-down submenu. Then, you will see a window shown in the next slide.

Move the variable [absingle] to the “Rows(s):” list box.

Move the variable [fundamen] to the “Columns(s):” list box.

Click on “Cells…” button on the bottom.

Mark the Column check box in the Percentages panel.

Click on the Continue button to close the dialog box.

Click on the OK button to produce the output.

There is a relationship in the bivariate table if the absolute value of the difference in column percentages on one of the rows is 10% or greater.

In this example, the difference is 40.8 - 51.2% = -10.4%.There is a relationship.

To add the control variable to the analysis, click on the Dialog Recall tool button and select the Crosstabs command from the pop-up menu.

To add the control variable to the analysis, move the [conserv] variable to the list box in the Layer panel.

Click on the OK button to produce the output.

The partial table for the first value or condition of the control variable (political orientation = not conservative) is in the top half of the three-way crosstabs table.

The partial table for the second value or condition of the control variable (political orientation = conservative) is in the bottom half of the three-way crosstabs table.

To compare column percentages, we compute the difference between the lower left-hand cell in the partial tables and the bivariate table

Comparing percentages in the lower left-hand cell of the tables, the percentage in both the first partial table (57.1%) and the second partial table (36.4%) are MORE than 5% different from the bivariate table (51.2%).

To compare column percentages, we compute the difference between the lower left-hand cell in the partial tables and the bivariate table

Comparing percentages in the lower right-hand cell of the tables, the percentage in both the first partial table (65.2%) and the second partial table (17.4%) are MORE than 5% different from the bivariate table (40.8%).

Elaboration of Bivariate Tables:Detecting a Conditional Relationship

With the information about the percentage differences in the partial tables and the bivariate table, we can answer the question.

Question: elaboration of bivariate table reinforces the finding of a relationship (a direct relationship)?

Is there a relationship between independent and dependent variables?

No

False

Yes

Are there relationships in both partial tables?

No

False

Yes

Are the differences in percentages less than 5%?

No

False

Yes

True

Question: elaboration of bivariate table negates the finding of a relationship (a spurious or intervening relationship)?

Is there a relationship between independent and dependent variables?

No

False

Yes

Are there relationships in both partial tables?

YES

False

NO

True

Question: elaboration of bivariate table alters the interpretation of a relationship (a conditional relationship)?

Is there a relationship between independent and dependent variables?

No

False

Yes

Are the cell percentages in partial tables different by 5% or more from bivariate table?

No

False

Yes

True