Chapter 1: Lecture Template (1 of 2)
I. Five Research Methods – method selected affects your findings
1. Case Study – a method that involves obtaining an in-depth analysis of a single “case” – for example - a
single individual, school, private/public corporation or institution, a geographical place, etc. (n=1)
- Designed to “describe” better (really depends on research goal)
- Clinical setting (in past)
- Only study that reveals deep, rich details
- More popular now because we don’t focus solely on generalizing
Why are case studies considered problematic methods of conducting research (according to some)?
- Poor/Lack of generalizability of findings to others – not confident that what you find in one case applies
to other cases
- Not a well-respected method (people only point out problems)
2. Observational Study – a method that involves observing people, animals, or any phenomenon.
Naturalistic studies involve observing things in their natural or unaltered environment (to maximize
ecological validity – or – the authenticity of data).
- More work, usually isn’t first choice
- You show up and observe
- Very time consuming – have to sift through irrelevant footage
o E.g. Observing high school bullying – you cannot set up a situation to purposely cause bullying
Lab studies involve observing things in a controlled environment. Specifically, researchers set up the
environment to expose particular reactions/responses.
- Inviting people into an environment to expose something (e.g. bullying)
- Manipulating and intervening as a researcher
- Researcher is more in control (experiment is standardized)
- Worry: does the behavior reflect true responses?
3. Survey – a method that involves measuring participants’ responses to specific questions
designed or compiled by the researcher. Participants simply answer these questions (i.e.,
participants self-report – nice option, only if it fits the research). Once participants have
answered the questions, researchers may mathematically calculate frequencies, means, or
relationships between participants’ responses. If researchers elect to investigate whether
there is a relationship between two or more things measured in a survey, then they typically
calculate a correlation. Correlations measure the size of the relationship between two or
more continuous variables. For example, if students study more, do they get higher grades? What is the difference between a positive and negative correlation? – are 2 variables related?
Positive correlation – direct relationship between 2 variables; increase one, other will also increase
- E.g. time spent studying + GPA (increase studying, increase GPA)
Negative correlation – inverse relationship between 2 variables; increase one, the other decreases
What is the difference between a variable and a measure?
Variables – are measured; e.g. school achievement or learning
Measure – e.g. by (i) an instrument, (ii) observation, (iii) questions, test
- How you collect data, how are you going to measure?
- The actual question that you used to find out about school achievement
- Tool that you are using to measure the variable
- Operationalization – process of defining variables and determining how you’ll measure them in a study
4. Experiment – a method that involves testing a causal hypothesis or causal research question. For
example, does one thing (IV) have a causal effect on another thing (DV)? IV DV
- E.g. Want to figure out what causes bullying
Independent Variable (IV) – the presumed causal variable that is manipulated by the researcher.
- E.g. things that contribute to bullying (low self-esteem, no attention from parents, etc.)
Dependent Variable (DV) – the characteristic or behaviour that is (expected to be) affected by the IV
- E.g. bullying
In an experiment, researchers manipulate and then measure variables. For example, researchers manipulate
the IV and measure the DV. In a survey, researchers measure variables only. In an observational study
(naturalistic or in a lab), researchers measure variables only
How do researchers “manipulate the IV”?
- E.g. using drugs to lower cholesterol, increase serotonin, etc. (pills)
- Researchers do something to manipulate the variable (problem)
- Manipulating to see what the effect is
- Workshop (altering someone’s knowledge) – educational research
- Adjunct questions learning DV
- “treatment” – created a software program with adjunct questions embedded.
o Software program (bury treatment in program)
* Key point is that you are manipulating – only manipulate IV (either increase or decrease)
5. QuasiExperiment – like an experiment, a quasi-experiment enables a researcher to test a causal hypothesis/research question. In addition, a quasi-experiment involves manipulation of the IV. However,
there is one difference between an experiment and a quasi-experiment…
What is the difference between an experiment and a quasi-experiment?
- Whether or not participants are randomly assigned to treatment and control groups
- If randomly assigned, then it is a EXPERIMENT
- If not randomly assigned, then it is a QUASI-EXPERIMENT
What is the difference between random assignment and random selection?
Random assignment – to treatment/control groups, deals with who gets the treatment and who doesn’t
Random selection/ random sampling – randomly selecting participants from population
- E.g. randomly selecting a school to randomly assign treatment/control
Is random selection necessary in an experiment or a quasi-experiment?
- NO, not necessary
- Experiment involves random assignment, not random selection
II. Practice Questions to Check your Understanding (complete after the lec