PSY 0035L:
Research Methods Lab
Name:______________________________________
I
N
C
LASS
E
XERCISE
#3
S
TATISTICS
R
EVIEW
O
BJECTIVES
The purpose of these exercises is to review various statistical tests that will important as you begin to
design your own experiments in the next couple of weeks.
This review covers:
•
the difference between descriptive and inferential statistics
•
median, mean (
M)
, mode, and standard deviation (
SD
)
•
the importance of statistical significance and its relationship to probability
•
the
t
test
•
the
chi square
(χ
2
) statistic
As part of the exercises you will be asked to calculate by hand statistics such as
M,
median
,
mode,
t,
and
χ
2
. You will also be introduced to the appropriate way of reporting statistical
results in APA style.
D
ESCRIPTIVE
AND
I
NFERENTIAL
S
TATISTICS
Descriptive statistics
summarize data.
For example, suppose you have the scores on a
standardized test for 500 subjects.
One way to summarize the data is to calculate a measure of
central
tendency
(
mean, median
, or
mode
) which indicates how the typical person scored.
You might also
determine the highest and lowest scores, and the spread of a distribution which would indicate how
much the scores varied (range and standard deviation).
Inferential statistics
are tools that tell us how much confidence we can have when we
generalize from a sample to a population.
You are familiar with national opinion polls in which a
carefully drawn sample of only about 1,500 adults is used to estimate the opinions of the entire adult
population of the United States.
The pollster first calculates descriptive statistics, such as the
percentage of respondents who are in favor of capital punishment and the percentage who are opposed.
Having sampled, he or she knows that the results may not be accurate because the sample may not be
representative; in fact the pollster knows that there is a high probability that the results are off by at
least a small amount.
This is why pollsters often mention a
margin of error
, which is an inferential
statistic. It is reported as a warning to the audience that random sampling may have produced errors,
which should be considered when interpreting results.
For example, a weekly news magazine reported
that in a national poll 58% of the respondents believed that the economy was improving; a footnote
indicated that the margin of error was ±2.3.
This means that the pollster was confident that the true
percentage for the whole population was within 2.3 percentage points of 58%.
You probably recall that a
population
is any group in which a researcher is interested.
It may
be large, such as all adults age 18 and over who reside in the United States, or it might be small, such
as all registered nurses employed by a specific hospital.
Researchers are free to choose populations of
interest and should clearly define them when writing reports of their studies.
A study in which all
members of a population are included is called a census.
A census is often feasible and desirable when
working with small populations (e.g., an algebra teacher may wish to pretest all students at the
Page 1 of 17
Revised 9/30/09
This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
This is the end of the preview.
Sign up
to
access the rest of the document.
 Spring '08
 BarbKucinski
 Statistics, Standard Deviation, Statistical hypothesis testing, Statistical significance, Research Methods Lab

Click to edit the document details