Exercise 13

 

To answer this question, we employ the goodness-of-fit chi square test. First of all, enter the data in the computer. Use 1 for those choosing psychology as a major, 2 for those choosing history and 3 for those choosing sociology; enter 15 times the digit 1, 11 times the digit 2, and 12 times the digit 3, all in one column. Having set the data in the computer, you follow the steps presented in the text (p. 384). The result obtained is shown below.

 

Test Statistics

 

	Subject choice
Chi-Square(a)	.684
df	2
Asymp. Sig.	.710

The important figure to consider here is .710, which indicates that the differences in the choice of study directions among the students considered in the study are not significant. Remember, significance is accepted if the level is .01 or smaller.

 

 

Exercise 14

 

To answer these questions we employ a chi square test. First of all, the tabular data must be entered in the computer. We demonstrated earlier how this is accomplished. The variables are 'couples' (1 for same-sex and 2 for other-sex couples) and 'comitmnt' (1 for high commitment and 2 for low commitment). The extra variable 'count' is set to accommodate the figures. Having set the data as required, we proceed as follows:

Go to Analyse/Descriptive Statistics/Crosstabs.

Transfer 'couples' to the top row and 'comitmnt' to the bottom row.

Go to Data in the menus and click Weight cases.

In the new dialog box activate Weight cases by clicking on the button in front of it.

Transfer 'count' to the Frequency Variable box, by clicking on the triangle in front of it.

Click OK, and in the new dialog box click Statistics and activate Chi square.

Click Continue and then OK to initiate processing.

Following this, you obtain the following table:

 

Chi-Square Tests

 

	Value	df	Asymp. Sig. (2-sided)	Exact Sig. (2-sided)	Exact Sig. (1-sided)
Pearson Chi-Square	1.220b	1	.269
Continuity Correctiona	.934	1	.334
Likelihood Ratio	1.221	1	.269
Fisher's Exact Test				.334	.167
Linear-by-Linear Association	1.215	1	.270
N of Valid Cases	214

a Computed only for a 2x2 table

b 0 cells (.0%) have expected count less than 5. The minimum expected count is 46.00.

 

 

Examining the significance level (.269) we can conclude that there is no evidence to suggest that the degree of commitment of the two groups of couples varies significantly.

 

 

Exercise 15

 

We employ exactly the same procedure followed in Exercise 14, with the exception that we activate Phi and Cramer's V. The results produced, using this procedure, are shown below.

 

Symmetric Measures

 

		Value	Approx. Sig.
Nominal by Nominal	Phi	.076	.269
	Cramer's V	.076	.269
N of Valid Cases		214

a Not assuming the null hypothesis.

b Using the asymptotic standard error assuming the null hypothesis.

 

As in the previous question, the important figure here is 'Approx. Sig.', the significance level, which is not only the same for both measures but also the same as that produced by the chi square test reported in the previous question.

 

 

Exercise 16

 

To answer the research question we employ the one-sample t-test. Here we set the scores in one column. We name the variable 'Satisfac' (for satisfaction). Remember that we are comparing means here. Further, we proceed as follows:

Select Analyze/Compare means/One-sample test.

Transfer 'satisfac' to the Test variable(s) box.

Type 6.2 in the Test value box.

Click Options and set Confidence interval to 95 per cent.

Activate Exclude cases analysis by analysis (in Missing values sector).

Click Continue and then OK.

 

Following this we obtain the following figures:

 

One-Sample Test

 

	Test Value = 6.2
	t	df	Sig. (2-tailed)	Mean Difference	95% Confidence Interval of the Difference
					Lower	Upper
Women's satisfaction	.764	24	.452	.152	-.26	.56

 

 

Of all data, the indicator of the significance level (.452) is the one we need to concentrate on. Given that this is greatly above the acceptable level of .05, we can conclude that the difference between the recorded scores and the average marital satisfaction is not significant.

 

 

Exercise 17

 

Given the nature of data and the level of measurement, as well as the relationship between the samples (they are paired), the most appropriate test to employ is the t-test for paired samples. After data entry, and using the variable names Flyflex and placebo, we proceed with the computation following the steps introduced in the text. The answer to the question is shown in the following table.

 

Paired Samples Test

 

		Paired Differences					t	df	Sig. (2-tailed)
		Mean	Std. Dev.	Std. Error Mean	95% Confidence Interval of the Difference
					Lower	Upper
Pair 1	Using Flyflex Using placebo	-1.00	.289	.058	-1.12	-.881	-17.32	24	.000

 

The level of significance is .000, which is the highest level of significance one can get. This indicates that the differences in the test performance of the two groups of respondents are significant. In more general terms, it means that taking Flyflex produces significantly different (and here, lower) anxiety scores than taking a placebo.

 

 

Exercise 18

 

Given the nature of the data and the number of samples, One-way ANOVA is the most appropriate test that could address the issues included in the question most satisfactorily. The variables are 'Achievemt' (dependent variable) and 'Residenc' (independent variable or factor). The residence is numbered 1 (city students), 2 (town students) and 3 (remote students). Set each student's score in one column, followed by the student's residence in the second column. The steps to take from here are those given in the text, namely

Select Analyze/Compare means/One Way ANOVA.

Transfer 'achievemt' to the Dependence List box.

Transfer 'residenc' to the Factor box.

Click Options and then Descriptive, then Continue and OK.

 

ANOVA

 

Achievement score

	Sum of Squares	df	Mean Square	F	Sig.
Between Groups	39.433	2	19.717	10.450	.000
Within Groups	107.550	57	1.887
Total	146.983	59

 

As in the previous exercise, so here, the level of significance is .000. This indicates that the differences in the test performance of the three groups of respondents are significant. This means that scholastic achievement is significantly affected by the students' place of residence.