Class Notes (1,100,000)
CA (620,000)
U of A (10,000)
PSYCO (1,000)
Lecture 2

PSYCO347 Lecture Notes - Lecture 2: Random Assignment, Experiment, Statistical Significance

Course Code
Jennifer Passey

This preview shows pages 1-2. to view the full 6 pages of the document.
Chapter 2: Research methods
Formulating a testable question:
-before any investigation can begin, the researcher must first generate a hypothesis (a testable
prediction about the relationship between two or more variables
-if there isn’t enough existing info to generate the hypothesis, the researcher will instead propose
a more general research question that clearly specifies the variables of interest but does not make
specific predictions about the associations among them
-some studies include both a hypothesis and research question, these can stem from a
researcher’s personal experiences
-most hypotheses and research questions are more commonly derived from existing research and
-essential feature for both is that it is testable
Defining variables:
-variables must be specified that are being explored
-before and predictions about these variables can be tested, each variable must be precisely
defined or operationalized
Operational definition- is an exact, specific and concrete definition of how the researcher chose
to conceptualize and measure the variable. Eg. Taking the variable marital satisfaction and
defining it in different ways, like overall happiness with their spouse on a 7-point scale.
-can operationalize marital satisfaction as scores on a multi-item questionnaire , multi-
operationalize (employ multiple measures of the variables included in their studies and increases
probability the variable’s meaning will be captured)
Selecting a research strategy:
Research strategy- concerns the way in which the researcher decides to investigate a question or
hypothesis, adopt one of three different research strategies
When the goal is to describe:
Descriptive research- no variables are manipulated and the researcher often has not set out to
specifically investigate a particular relationship among variables; rather, he or she is simply
interested in describing people’s behavior, free response data
-different types of descriptive research

Only pages 1-2 are available for preview. Some parts have been intentionally blurred.

Naturalistic or observational studies- involve detailed observation of a participant sample in a
natural or real-world setting
Survey research- involves designing and administrating a questionnaire to a sample of people in
order to assess and describe their beliefs, attitudes, behaviors, and so forth
Archival research- involves using data that were gathered by others. Public archives and
statistical abstracts provide data on marriage, divorce, births and deaths, and other important
interpersonal events
-descriptive studies often serve as a springboard for future research because they provide basic
info about the phenomenon of interest
-naturalistic/observational studies often have high levels of external validity (the results are
reflective of what actually occurs in the “real world”), however the researcher who conducts a
naturalistic study has no control over the participant or the research situation and therefore
cannot make causal inferences, time-consuming and challenging
-researchers who conduct surveys must decide how to obtain an appropriate sample of
participants and how to maximize response rates so that no biases will be introduced into their
results, constructing the questionnaire is also of importance in wording and the types of response
-archival research has difficulties in being extremely laborious, and data can be limited
When the goal is to establish association:
-see if two or more variables are related to each other in some meaningful way
Correlational research- involves measuring variables and looking for evidence of an association
between them. This does not allow scientists to determine whether one variable causes another,
but simply to determine whether the variables covary. This is useful for when the variables
cannot be directly manipulated or controlled
-Three types of correlations or associations that variables can demonstrate: positive, negative,
and none (zero)
Positive correlation- between two variables means that they covary in a positive way; as scores
on one variable (x) increase, so do scores on the other variable (y)
Negative correlation- between two variables means that they covary in a negative way; as scores
on one variable (x) increase, scores on the other variable (y) decrease
Zero correlation- means that the two variables are unrelated; there is no evidence of a significant
association between them
You're Reading a Preview

Unlock to view full version