CHAPTER 1: Research Methods
CONCEPT MAP
I. Introduction
II. Making Observations
A. Defining the Question
i. Testable Hypotheses—create an hypothesis that is testable
ii. Operational Definition—present definitions that are observable
and testable
iii. Construct Validity—create valid and true measurements
B. Parts of an Experiment
i. Independent Variables—what is manipulated by the experimenter
ii. Dependent Variables—what is measured by the experimenter
C. Systematically Collecting Data
i. Confirmation Bias—biases common to experiments may alter results
ii. Anecdotal Information—anecdotal information and common sense are
not scientific but may provide clues to important questions
D. Defining the Sample
i. Population—all possible representatives of a group
ii. Sample—a subset of the population
iii. Random Sample—several methods including maximum variation
, sampling used to ensure randomness to a sample and maximum
variant sampling
iv. Case Studies—single subject or limited case studies may also be used
E. Issues That May Compromise Experiments
i. External Validity—ensure that your study measures what it was
supposed to measure
ii. Demand Characteristics—be alert to characteristics of the
experimenter
or subject that hinder experiments
iii. Double-blind Design—reduce or eliminate experimenter and subject
bias
,III. Working with the Data
A. Descriptive Statistics
i. Frequency Distributions—tallying and tabling the data
ii. Measure of Central Tendency—ways to summarize and present the
data
iii. Finding the “Average”—means, median, modes
iv. Measures of Variability—characterizing variability of the data set
and summarizing variability using standard deviations
B. Correlations, Reliability, and Validity
i. Scatter Plots—graphing the data
ii. Correlation Coefficients—studying relationships between variables
iii. Reliability—measures of reliability such as test–retest reliability
and interobserver reliability
C. Inferential Statistics and Testing Differences
i. Testing Differences—determining whether experimental manipulation
produced an effect
ii. Statistical Significance—defining p-values and determining effect size
IV. Observational Studies: Alternatives to Experiments
A. Quasi-experiments
i. Observational Studies—studies that do not explicitly manipulate
variables
ii. Quasi-experimental Design—study of nonrandomly assigned variables
such as gender or age
, iii. Third Variable Problem—may be a third, unaccounted for, variable
that
is the causal agent for your results
B. Establishing Cause and Effect
i. Experimental Groups—the group that receives the manipulation
ii. Control Group—the group that gets no manipulation
iii. Random Assignment—assures that the experimental and control
group
are similar
iv. Within-Subject Comparisons—comparing a subject to itself before
and after a manipulation.
v. Power of Experiments—experimental design increases the power
of an experiment
CONCEPT MAP
I. Introduction
II. Making Observations
A. Defining the Question
i. Testable Hypotheses—create an hypothesis that is testable
ii. Operational Definition—present definitions that are observable
and testable
iii. Construct Validity—create valid and true measurements
B. Parts of an Experiment
i. Independent Variables—what is manipulated by the experimenter
ii. Dependent Variables—what is measured by the experimenter
C. Systematically Collecting Data
i. Confirmation Bias—biases common to experiments may alter results
ii. Anecdotal Information—anecdotal information and common sense are
not scientific but may provide clues to important questions
D. Defining the Sample
i. Population—all possible representatives of a group
ii. Sample—a subset of the population
iii. Random Sample—several methods including maximum variation
, sampling used to ensure randomness to a sample and maximum
variant sampling
iv. Case Studies—single subject or limited case studies may also be used
E. Issues That May Compromise Experiments
i. External Validity—ensure that your study measures what it was
supposed to measure
ii. Demand Characteristics—be alert to characteristics of the
experimenter
or subject that hinder experiments
iii. Double-blind Design—reduce or eliminate experimenter and subject
bias
,III. Working with the Data
A. Descriptive Statistics
i. Frequency Distributions—tallying and tabling the data
ii. Measure of Central Tendency—ways to summarize and present the
data
iii. Finding the “Average”—means, median, modes
iv. Measures of Variability—characterizing variability of the data set
and summarizing variability using standard deviations
B. Correlations, Reliability, and Validity
i. Scatter Plots—graphing the data
ii. Correlation Coefficients—studying relationships between variables
iii. Reliability—measures of reliability such as test–retest reliability
and interobserver reliability
C. Inferential Statistics and Testing Differences
i. Testing Differences—determining whether experimental manipulation
produced an effect
ii. Statistical Significance—defining p-values and determining effect size
IV. Observational Studies: Alternatives to Experiments
A. Quasi-experiments
i. Observational Studies—studies that do not explicitly manipulate
variables
ii. Quasi-experimental Design—study of nonrandomly assigned variables
such as gender or age
, iii. Third Variable Problem—may be a third, unaccounted for, variable
that
is the causal agent for your results
B. Establishing Cause and Effect
i. Experimental Groups—the group that receives the manipulation
ii. Control Group—the group that gets no manipulation
iii. Random Assignment—assures that the experimental and control
group
are similar
iv. Within-Subject Comparisons—comparing a subject to itself before
and after a manipulation.
v. Power of Experiments—experimental design increases the power
of an experiment
