Statistical Methods for Psychology

other scoring low. When presented with sexual explicit homosexual videos, the homophobic

group actually showed a higher level of sexual arousal (the mean difference 5 7.50 units). A

ttest of the difference between means produced a statistically significant result (p,.05), and

Cohen’sd 5 .62 showed that the two groups differed by nearly 2/3 of a standard deviation.

However, the confidence limits on the population mean difference were rather wide (1.46

m 1 – m 2 13.54, suggesting that we do not have a tight handle on the size of our difference.

SPSS Analysis

The SPSS analysis of the Adams et al. (1996) data is given in Table 7.6. Notice that SPSS

first provides what it calls Levene’s test for equality of variances. We will discuss this test

shortly, but it is simply a test on our assumption of homogeneity of variance. We do not

come close to rejecting the null hypothesis that the variances are homogeneous (p 5 .534),

so we don’t have to worry about that here. We will assume equal variances, and will focus

on the next-to-bottom row of the table.

Next note that the tsupplied by SPSS is the same as we calculated, and that the proba-

bility associated with this value of t(.016) is less than a 5 .05, leading to rejection of the

null hypothesis. Note also that SPSS prints the difference between the means and the stan-

dard error of that difference, both of which we have seen in our own calculations. Finally,

SPSS prints the 95% confidence interval on the difference between means, and it agrees

with ours.

7.6 A Second Worked Example

Joshua Aronson has done extensive work on what he refers to as “stereotype threat,” which

refers to the fact that “members of stereotyped groups often feel extra pressure in situations

where their behavior can confirm the negative reputation that their group lacks a valued

...

...

Section 7.6 A Second Worked Example 211

N

35

29

GROUP

Arousal Homophobic

Nonhomophobic

Mean

24.0000

16.5034

Std.

Deviation

12.2013

11.7966

Std. Error

Mean

2.0624

2.1906

Group Statistics

Independent Samples Test

Equal variances

assumed

Equal variances

not assumed

F

.391

Sig.

.534

t

2.484

2.492

df

62

60.495

Sig.

(2-tailed)

.016

.015

Mean

Difference

7.4966

7.4966

Std. Error

Difference

3.0183

3.0087

Lower

1.4630

1.4794

Upper

13.5301

13.5138

95% Confidence

Interval of the

Difference

t Test for Equality of Means

Levene’s Test

for Equality of

Variances

Table 7.6 SPSS analyses of Adams et al. (1996) data