Chapter 1 Research Methods
Variable - any characteristic whose values can change.
Defining the Question
We are guided by the questions we hope to address through our observations.
Not enough to just have a question. The question must be formulated in a way that the evidence can
link to it.
Testible Hypothesis: a prediction that has been formulated specifically enough so that it is clear what
observations would confirm the prediction and what observations would challenge it.
This has a specfic claim about the facts, framed in a way to allow an unambiguous test.
Testing must make hypothesis falsifiable (stated so that we're clear at the start about what
pattern in the evidence could show the hypothesis to be false.
Operational Definition: A definition that translates the variable we want to assess into a specific
procedure or measurement.
Construct Validity: must truly reflect variable names in hypothesis.
Dependent Variable: variable measured or recorded in an experiment.
Independent Variable: variable the experimenter manipulates as a basis for making predictions about
the dependent variable.
Systematically Collecting Data
Anecdotal Evidence: evidence that involves just one or two cases, has been informally collected, and is
now informally reported.
Defining the Sample
Population: the entire group about which the investigator wants to draw a conclusion.
Sample: a subset of the population they are interested in.
Random Sampling: procedure in which every member of the population has an equal chance of being
picked to participate in a study.
Maximum Variation Sampling: stratedgy of delibrately seeking out the unusual or extreme cases.
Cast Study: instensive study of one person in great detail. Case studies play an important role in the study of the brain
Assesing External Validity
External Validity: degree to which a study's participants, stimuli, and procedures adequately reflect the
world as it actually is.
Sample of people in study need to be representative of broader population.
Circumstances of study need to reflect those in the broader world.
Depends on what's being investigated.
Questions of external validity need to be resolved through research
Monitoring Demand Characteristics
Those who know they are being observed may change their behavior because of it.
Demand Characteristics: the cues in study that might tell a research participant what behaviors are
expected or desirable in that setting.
can influence an entire data pattern
can create artificial differences between the groups being compared.
Double Blind Design: technique of assigning participants to experimental conditions while keeping both
the participants and the researchers unaware of who is assigned to which group.
Working With Data
After defining a question, choosing a sample, and collecting data, we have Statisical Analysis of Data.
2 parts. 1st researchers rely on Descriptive Statistics (mathematical procedures that allow a
researcher to characterize a data pattern; these prodecures include measures of central
tendency and of variability) to summerize the data. 2nd researchers will us inferential statistics
(mathematical procedures that allow a researcher to draw further claims from a data pattern,
including claims about whether the pattern observed in the sample is likely to be in other
samples) to ask how confident the can be in drawing conclusions based on their sample.
At this point researchers should know if their data supports their hypothesis or not.
Frequency Distribution: a table that lists how many scores fall into each of the designated categories.
Means and Variability
Large number of moderate values and fewer and fewer as we move away from the center.
1st locate curve's center. Gives us measure of the "average case" within data set (AKA measure of
central tendency). Mean: determines the average. Add up all the scores and then divide the sum by the number of tests
Median: put data values in order and findig the value that divides the distribution in half.
Variability: degree to which scores in a frequency distribution depart from the central value. How much
individual scores vary from one to the next.
Standard Deviation: a measure of the variability of a data set, calculated as the square root of th