unit 2 exam Flashcards
what is a true experiement?
-researcher manipulates an independent variable
-tests whether the manipulation influences a dependent variable
what is a correlational study?
-researcher does not manipulate independent variable
-examines relationship b/w variables they chose to examine
what is an Ex Post Facto?
-special case of correlational study
-researcher compares groups of individuals who differed prior to their participation of in the study
what is a variable?
-something that varies, if participants are all the same it is not a variable
what is true about variables?
-the levels of a variable are the different specific possibilities (ex. age, reaction time)
-each variable has at least two possibilities/alternatives
what do levels represent in terms of variables?
number of alternatives being examined or manipulated in the study (ex. high noise vs low noise = 2 levels)
what is an independent variable?
-the variable being manipulated
-experimenter manipulated (true experiment)
-experimenter selected (subject variable, ex post facto)
what is a dependent variable?
variable whose reaction is being observed/measured
what is true about correlational designs?
it is not clear which is the independent and dependent variables
what is an operational definition?
a statement of the operations necessary to produce and measure a concept or construct
true or false: all variables in psychological research must be operationally defined.
true
what are these examples of?
-people with panic disorder vs. people with no psychiatric disorder
-individuals in noise condition vs. individuals in quiet conditon
contrast
true or false: you should attempt to draw an inference (conclusion) on the basis of the contrast
true
in an experiment, how should you hold conditions constant?
control for everything except the independent variable about which you want to draw an inference
what is an extraneous/confounding variable?
any variable you are not interested in that is not controlled for
what are the two types of controls?
1.between-subject design: there is an experimental and a control group
2.within-subject design: subjects serve as own control by comparing a single group of individuals under two+ different conditions
what is a placebo?
-an intervention that is presented to the participant as though it is the real intervention that’s expected to have an impact
can there be a placebo in psychotherapy research?
it is unclear because part of impact is attention from a therapist
what is experimenter bias?
the results of a study being influenced by the experimenter’s unconscious belies/biases
what are demand characteristics?
subtle cues in a research study that can unintentionally signal to participants what the researcher is expecting of them or hoping to find
what can be used to limit or prevent experimenter bias and demand characteristics?
placebos
what is block randomization/random blocks technique?
rank order all participants, take groups of individuals with adjacent rank ordered scores and then randomly assign groups
which type of group membership assignment poses the greatest risk of group differences due to bad luck?
random assignment
which type of group membership assignment poses the smallest risk of group differences due to bad luck?
matching
which type of group membership assignment poses an intermediate risk of group differences due to bad luck?
block randomization
which type of control group has an advantage for the possibility of group differences?
within subject control
which type of control group has an advantage for order and practice effects?
between subjects
psychological research typically wishes to learn about…
populations
what is a sample?
a subset of the population about which you wish to draw conclusions
How confident one can be generating conclusions
about populations based on research conducted
with samples depends on:
-How representative is the sample (is the sample
systematically unrepresentative – referred to as “biased”)
-The size of the sample
what did the berenbaum (2008) study try to do?
attempted to obtain random, representative sample of geographical region in which study was conducted
what is random digit dialing?
purchase phone numbers in area you are studying
what is descriptive statistics?
intended to do nothing more than describe (ex. average age is participants in sample is 38)
what are inferential statistics?
intended to enable researchers to draw inferences (usually involves probabilities)
why are inferential statistics important?
-used to determine how confident one can be about the sort of inferences/conclusion one wishes to draw
-need them to draw inferences about populations when measuring samples
what are the four levels of measurement?
1.Nominal: categorical
2.Ordinal: numerical, relative rank ordering, intervals b/w successive values are not always equal (ex. never, sometimes, rarely)
3.Interval: numerical, intervals b/w successive values are equal, no true zero, ratio of different numbers is not meaningful (ex. degrees Fahrenheit)
4.Ratio: numerical, intervals b/w successive values are equal, has true zero, ratio of different numbers is meaningful (ex. length, mass)
what does quality of research depend on?
-how confident you can be about your inference (more confident = better quality)
-how specific or detailed an inference you can draw (more detailed = better quality)
what is internal validity?
-degree to which the experimenter’s manipulation of the independent variables can account for the results (did the experimenter do what they said they were gonna do)
-degree to which the results can be definitively attributed to the effects of the independent variables
historically, what have been some potential threats to internal validity?
-unplanned events experienced by research participants (especially for longitudinal studies)
-ex: economy improves and unemployed participants get jobs
what is maturation?
-changes that take place within research participants (relevant to studies that take place over an extended period of time)
what is testing?
the effect of testing at one point in time on taking a similar or identical test at a different point in time
what is instrumentation/measurement ?
-the effect of changes in the instruments used or the way things are measured
-ex: Some participants’ levels of anxiety measured by an inattentive, insensitive judge whereas other
participants’ levels of anxiety measured by an
attentive, sensitive judge.
what is statistical regression?
the tendency for extreme scores to regress toward the mean when the test/measure is readministered
what is selection bias?
when the selection or assignment of participants to groups leads the groups to differ in some way
what is attrition?
when not all participants complete the study
what is diffusion/imitation of treatment?
relevant only to treatment studies when individuals in a control group unintentionally receive some or all of the treatment being provided to individuals in the experimental group
what is external validity?
the degree to which the results can be generalized from a given study to other organisms, people, times, places, and measures
what is the concern for external validity when it comes to stimulus characteristics and settings?
-will the results generalize when different aspects of the study that are not supposed to be crucial (like setting, interviewer, experimenter) are changed?
-will results generalize outside of lab setting?
what is meant by the reactivity of experimental arrangements?
-are the results influenced by the fact that the participants know that they are being studied?
what is meant by pretest sensitization?
-are the results influenced by testing/assessment prior to the manipulation/intervention
-would the results generalize if a pretest were not administered
what is meant by timing of measurement?
-to what degree are the results a reflection of the specific point in time when assessment occurred
testing participants under highly controlled lab conditions helps to maximize/reduce which type of validity?
internal maximized
external reduced
Examining participants under the most natural conditions possible maximizes/reduces what kind of validity?
external maximized
internal reduced
True or false: as internal validity increases, external validity decreases and vice versa.
True
what is correlation?
quantitative measure of how strongly two variables are associated
what does a correlation of 0 mean?
no association
true or false: the greater the absolute value of the correlation, the stronger the association
true
what kind of association is this: as scores on one variable increase, scores on the other variable increase
positive
what kind of association is this: as scores on one variable increase, scores on the other variable decrease
negative
what is the most common way correlation is measured?
Pearson product-moment
what is the Pearson product-moment?
-measures the strength of a linear relationship b/w two variables
-more strongly influenced by outliers
what is the spearman rank-order?
-takes into account the rank order of the score, not the actual raw score
-outliers have less impact
what is a scatterplot?
a diagram using cartesian (X and Y) coordinated to display values for two variables of a data set
what is a correlational design?
-researcher does not manipulate any variables
-simply measures two+ variables and examines how they are associated
which kind of correlational studies examine association b/w a categorical variable and a dimensional variable?
ex post facto studies
which is the categorical variable and the dimensional variable in this scenario: Do people with OCD have poorer memory than non-psychiatric controls?
-Categorical Variable: OCD vs. Non-psychiatric control
-Dimensional Variable: score on memory test
what are three possible causal reasons for two variables to be correlated?
-A causes B
-B causes A
-Some other variable (C) causes both A and B
what is unique about ex post facto studies?
-researcher selects, instead of manipulates, independent variable
true or false: in ex post facto studies, the groups differ in multiple ways in addition to diagnostic status
true
what is the biggest problem with ex post facto design?
-group differences
how can researchers deal with group differnces?
-can try to compensate for group differences by matching groups on natural differences like race, gender, or education
what is the problem with matching in ex post facto designs?
-when two groups match on one variable, they may now mismatch on a second variable that they otherwise would not have differed on
-at least one of the groups will no longer be representative of the population they came from
what is mean?
average
what is median?
score located in the center of distribution (useful when there is outliers)
what is a histogram?
a display of statistical info that uses rectangles to show frequency of data items
what does the height (y axis) of each rectangle (bar) indicate?
the frequency of scores
where are scores typically indicated in a histogram?
on the x axis
what is an outlier?
a score that deviates from the vast majority of scores
what do outliers to the mean?
pull the mean away from the middle of the distribution and toward the outlier (s)
what is range?
-how spread apart scores are
-the difference between the lowest and highest scores
what is standard deviation?
how close scores tend to be to the mean
the smaller the standard deviation, the _____ the scores tend to be to the mean.
closer
the _______ the standard deviation, the further the scores tend to be to the mean
larger
are smaller standard deviation more or less homgenous?
more homogeneous
which type of distribution looks like a pyramid?
unimodal distribution
which type of distribution looks like two pyramids?
bimodal disitribution
which type of distribution looks like a bell curve?
symmetric/normal distribution
which type of distribution has no skew and has a mean that equals the median?
symmetric/normal distribution
which type of distribution has a skew when outliers are primarily on one end of the distribution?
asymmetric distribution
