Increase in the number of independent variables in...

Increasing the number of independent variables in the construction of cross-tabulation tables

You can build cross-tabulation tables by capturing the values ​​of several independent variables.

There are four possible situations [30].

First, sometimes adding one more independent variable clarifies the mechanism of the previously discovered dependency. So, by calculating the cross-tabulation table between the marital status (independent variable) and the level of acquiring fashionable clothing (Table 12.13), one would think that many men and women after marriage or marriage lose interest in buying fashionable clothes.

Table 12.13. Distribution of persons with different marital status by the number of fashionable clothes they acquire,%

Get fashionable clothes

Marital status

All respondents

Married

Single (not married)

Many

31

52

37

Not enough

69

48

63

Total

100 (700)

100 (300)

100 (1000)

However, the inclusion in the analysis of another independent variable - the sex of the respondent (Table 12.14) shows that this pattern is manifested only in women, and it is more pronounced than in the respondents as a whole.

Table 12.14. Distribution of persons of different sexes and with different marital status by the number of fashionable clothes they acquire,%

Acquire

fashionable clothing

Sex

All

Interviewed

Men

Women

Marital status

Marital status

Married

Single

Married

Single

Many

35

40

25

60

37

Not enough

65

60

75

40

63

Total

100 (400)

100 (120)

100 (300)

100 (180)

100 (1000)

Secondly, it sometimes turns out that the previously observed dependence was illusory, the so-called false correlation ; that in fact there is another factor, the variation of which explains the observed effects. So, the tab. 12.15 gives the impression that people with higher education often purchase expensive car brands.

Table 12.15. The presence of an expensive brand car in people with different education,%

The presence of an expensive car

Education

All

Interviewed

Higher

not higher

Yes

32

21

24

No

68

79

76

Total

100 (250)

100 (750)

100 (1000)

The inclusion in the analysis of another independent variable - the income of the respondent (Table 12.16) - shows that education in itself does not affect the probability of buying an expensive car; the true cause of the observed differences is the level of income that people with higher education tend to have higher.

Table 12.16. The presence of an expensive brand car in persons with different incomes and education,%

Availability

expensive

car

Revenue

All respondents

Low

high

Education

Education

Higher

not higher

Higher

not higher

Yes

20

20

40

40

No

80

80

60

60

Total

100

100

100

100

100

(100)

(700)

(150)

(50)

(1000)

Thirdly, sometimes adding one or more independent variables allows us to reveal a previously hidden dependence. For example, an attempt to identify the assumed link between age and interest in traveling abroad has failed (Table 12.17).

Table 12.17. Interest in overseas tourism in persons of different ages,%

Interest in overseas tourism

Age

All

Interviewed

up to 45 years old

45 years and older

Interested

50

50

50

Not interested

50

50

50

Total

100 (500)

100 (500)

100 (1000)

Dividing the same respondents by sex (Table 12.18), the researchers found the desired dependency, which in men and women was multidirectional.

Table 12.18. Interest in overseas tourism in persons of different sex and age,%

Interest in overseas tourism

Sex

All respondents

Men

Women

Age

Age

up to 45 years old

45 years and older

up to 45 years old

45 years and older

Interested

60

40

35

65

50

Not interested

40

60

65

35

50

Total

100

(300)

100

(300)

100

(200)

100

(200)

100

(1000)

Finally, fourthly, it is possible that the inclusion of independent variables in the analysis does not change anything in relation to the previously identified or, conversely, unidentified pairing pattern.

In general, increasing the number of independent variables when building cross-tabulation tables is useful. But you should not abuse this. It is impossible to allow the formation of so small groups in rows and columns in the analysis, so that the condition fe ≥ 5 is violated, where fe is the expected number of respondents in the cell of the cross-tabulation table, assuming that its rows and columns are independent.

thematic pictures

Also We Can Offer!

Other services that we offer

If you don’t see the necessary subject, paper type, or topic in our list of available services and examples, don’t worry! We have a number of other academic disciplines to suit the needs of anyone who visits this website looking for help.

How to ...

We made your life easier with putting together a big number of articles and guidelines on how to plan and write different types of assignments (Essay, Research Paper, Dissertation etc)