Scribd
Upload
236 views18 pages

Data Analysis/Interpretation: Describing Data, Confidence Intervals, Correlation

This document discusses data analysis and interpretation. It outlines three main steps: 1) Getting to know the data by examining features, looking for errors, and visualizing distributions 2) Summarizing the data using descriptive statistics like means, medians, modes, ranges and standard deviations 3) Confirming what the data reveal using confidence intervals to determine if observations support claims about behavior or mental processes. It provides examples of confidence intervals for single means, comparisons between independent groups, and repeated measures designs.

Uploaded by

crutili
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
236 views18 pages

Data Analysis/Interpretation: Describing Data, Confidence Intervals, Correlation

This document discusses data analysis and interpretation. It outlines three main steps: 1) Getting to know the data by examining features, looking for errors, and visualizing distributions 2) Summarizing the data using descriptive statistics like means, medians, modes, ranges and standard deviations 3) Confirming what the data reveal using confidence intervals to determine if observations support claims about behavior or mental processes. It provides examples of confidence intervals for single means, comparisons between independent groups, and repeated measures designs.

Uploaded by

crutili
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 18

Data Analysis/Interpretation:

Describing Data, Confidence


Intervals, Correlation
Week 8
10/28 and 10/30
Ch. 12
The Next Step
• So far we have learned about methodological
issues and how to collect data
• Now, we will talk about how to interpret the
data
• Primary goal of psychological research is to
make claims about behavior or mental
processes
– Primary goal of data analysis is to determine
whether our observations support our claim about
behavior/mental processes
Overview of Data Analysis
• Research methods (e.g., hypotheses,
study design, participants, etc.) affect
our data
– If errors/biases were made during data
collection, results can be hindered
– Might not be able to “save” data-- have to
start over
Stages of Data Analysis
• Getting to Know the Data
• Summarizing the Data
• Confirming What the Data Reveal
• Use confidence intervals
Step 1: Getting to Know the
Data
• Examine general features of data
• Look for errors in data-- missing data or
impossible numbers
– Edit or “clean” data
– Make corrections, decide what to do with
“anomalies” and outliers
• If data is removed, must report removal
• View visual displays-- distribution of data
– What is the variability/dispersion of data? Is it
“normal” or “skewed?” Should it be transformed?
Example of Outlier
Example of Normal
Distribution
40

35

30

25

20

15

10

0
1 2 3 4 5
Example of Skewed
Distribution
35

30

25

20

15

10

0
1 2 3 4 5
Step 2: Summarizing Data
• If data appears to “make sense” in Step
1, we proceed to Step 2
• Summarizing/describing the data
– Descriptive Statistics
– Info typically goes in Methods or beginning
of Results sections-- help describe what
our data “looks like” to the reader
Summarizing Data
continued...
• Measurements involved in summarizing data:
• Central Tendency
– Mean, median, mode
• Variability/dispersion
– Range, standard deviation
– Standard error of the mean
• Measures of effect size
– Strength of relationship/effect between IV’s and
DV’s
Step 3: Confirm What the
Data Reveal
• Use confidence interval (CI) for a population
parameter
• Margin of error-- our sample provides an
estimation of a population value; possibility
for error in our sample
• Usually with psychology use .95 probability (p
< .05)
– We can be 95% sure the value for the population
falls within these numbers/percentage (margin of
error)
Example of Confidence
Interval
CI continued...
• Different calculations of CI:
• CI for a Single Mean
• CI for Comparison Between
Independent Group Means
• CI for Comparison Between Means in a
Repeated Measures Design
CI for a Single Mean
• One variable/score
• 95% sure the CI contains true
population mean
– e.g., estimating GPA of all DePaul students
based on GPA of students in this class
– 95% sure it is between 3.1 and 3.4
CI for Comparison Between
Independent Group Means
• 95% sure CI contains true population mean
difference
– E.g., estimating GPA of Psychology majors versus
History majors, versus Chemistry majors
• Width of CI provides info about effect size
• Determine whether CI of mean difference
includes possible score of 0
– If so, there may be no population difference
CI for Comparison Between
Means in a Repeated
Measure Design
• Same sample/participants studied over
numerous trials
– E.g., DePaul students’ GPA when they were
freshmen, sophomores, juniors
• Nonoverlapping CI between groups/times
suggests sample means are not the same--
true difference
• Overlapping CI between groups/times--
uncertain whether there is a difference
CI for Comparison Between
Means in Repeated Measure
Design

QuickTimeª and a
TIFF (LZW) decompressor
are needed to see this picture.
Notes about CI
• You can change CI
– Higher the CI, the less specific your
estimate will be
– E.g., 95% CI versus 99% CI
• Need to include larger margin of error to
include possible values
• Increasing sample size will improve
estimate of mean (decreases standard
error of the mean)

You might also like