Social Research Methods: Qualitative and Quantitative Approaches

ANALYSIS OF QUANTITATIVE DATA

increased accuracy or clarity in public debate. More
often, it has increased confusion; this makes know-
ing what statistics can and cannot do essential. The
cliché that you can prove anything with statistics is
false; however, some people can and do misuse sta-
tistics to pretend to prove anything. Through igno-
rance or conscious deceit, some people use statistics
to fool others. The best way to protect yourself from
being misled by statistics is not to ignore them or

hide from the numbers but to understand the

research process and statistics, think about what you

hear, and ask questions.

We turn next to qualitative research. The logic

and purpose of qualitative research differ from those

of the quantitative, positivist approach of the past

chapters. It is less concerned with numbers, hypothe-

ses, and causality and more concerned with words,

norms and values, and meaning.

KEY TERMS

bivariate statistics
codebook
coding procedure
contingency cleaning
contingency table
control variable
covariation
cross-tabulation
curvilinear relationship
data field
data records
descriptive statistics
direct-entry method
elaboration paradigm
explanation pattern
frequency distribution

frequency polygon

histogram

inferential statistics

interpretation pattern

level of statistical

significance

linear relationship

marginal

mean

measures of central tendency

median

mode

net effect

normal distribution

partials

percentile

possible code cleaning

proportionate reduction in error

range

replication pattern

scattergram

skewed distribution

specification pattern

standard deviation

statistical independence

statistical relationship

statistical significance

suppressor variable pattern

Type I error

Type II error

univariate statistics

z-score

REVIEW QUESTIONS

1.What is a codebook, and how is it used in research?

2.How do researchers clean data and check their coding?

3.Describe how researchers use optical scan sheets.

4.In what ways can a researcher display frequency distribution information?

5.Describe the differences between mean, median, and mode.

6.What three features of a relationship can be seen from a scattergram?

7.What is a covariation, and how is it used?

8.When can a researcher generalize from a scattergram to a percentaged table to

find a relationship among variables?

9.Discuss the concept of control as it is used in trivariate analysis.

10.What does it mean to say “statistically significant at the .001 level,” and what type
of error is more likely, Type I or Type II?