Clinical Psychology

whereas for those with more severe problems, treat-
ment 2 would be preferable.
Later in this book, it will become apparent that
there is no“best”therapy for all problems and all
people. There are only treatments that vary in their
effectiveness for different kinds of psychological
problems and different kinds of people. Mixed
designs can help us discern what is best for whom.
Of course, we must not forget that in mixed
designs, one of the factors (e.g., severity of illness)
is not manipulated, and this raises the kinds of
problems discussed earlier in the case of correla-
tional methods (Davison et al., 2004).

Strengths and Weaknesses of Research Methods

As we have surveyed the various research methods
commonly used by clinical psychologists, a number
of strengths and weaknesses have been highlighted
for each. Table 4-4 summarizes major features of
these approaches, focusing on each method’s ability
to provide detailed information about a single case,
to exercise control over extraneous factors that
might influence the effect of interest (i.e., internal
validity), to generalize to other people or other
situations, and to make causal inferences. For exam-
ple, the experimental method, the single-case
design, and the mixed design all have high internal
validity. However, only the experimental method

and single-case design allow an investigator to make

causal inferences.

Statistical Versus Practical Significance

After a statistic (e.g., a correlation coefficient) has

been calculated, it can be determined whether the

obtained number is significant. Traditionally, if it is

found that the obtained value (or a more extreme

value) could be expected to occur by chance alone

less than 5 times out of 100 (i.e., very rarely), it is

deemed statistically significant. Such an obtained

value is said to be significant at the .05 level, usually

written asp< .05. The larger the correlation, the

more likely it is to be significant. But when large

numbers of participants are involved, even relatively

small correlations can be significant. With 180 par-

ticipants, a correlation of .19 will be significant;

when only 30 participants are involved, a correlation

of .30 would fail to be significant.

Therefore, it is important to distinguish

betweenstatistical significanceandpractical significance

when interpreting statistical results. The correlation

of .19 may be significant, but the magnitude of the

relationship is still quite modest. For example, it

might be true that in a study involving 5,000

second-year graduate students in clinical psychology

T A B L E 4-4 Advantages and Limitations of Major Research Methods Used
in Clinical Psychology

Case Study

Methods

Epidemiology/

Correlational

Methods

Experimental

Methods

Single-Case

Designs

Mixed

Designs

Detailed
information
Internally valid
results
Externally valid
results

Can determine
causality

118 CHAPTER 4