RESIT research & statistics Flashcards
inference
making decisions and predictions based on the data for answering the statistical question
inferential statistics
methods of making decisions or predictions about a population, based on data obtained from a sample of that population
parameter
numerical summary of the population
categorical variable
if each observation belongs to one of a set of categories
quantitative
if observations on it take numerical values that represent different magnitudes of the variable
discrete quantitative variable
0,1,2,3,4, if its possible values form a set of separate numbers
continuous quantitative variable
if its possible values form an interval
modal category
category with the highest frequency (for a categorical variable)
mode
(for quantitative variable), the numerical value that occurs most frequently
unimodal
data has single mount
bimodal
data has two distinct mounds
median
middle value of the observations when the observations are ordered from smallest to the largest(or other way around)
resistent
if extreme observations have little if any influence on its value
deviation
of an observation x from the mean , the difference between the observation and the sample mean
variance
average of the squared deviations
pth percentile
value such that p percent of the observations fall below or at that value
z-score
for an observation is the number of standard deviations that it falls from the mean. positive z score indicates the observation is above the mean, negative below the mean
empiricism
involves using evidence from the senses or from instruments that assist the senses as the basis for conclusions.
parsimony
theories are supposed to be simple. mid two theories explain the at a equally well, most scientists will opt for the ampler more parsimonious theory
applied research
conducting in real world context
probabilistic
behaviour research findings are not expected to explain all cases all of the time.
availabiliity heuristic
things pop up easily in our mind and to guide our thinking
present bias
name for our failure to consider appropriate comparison groups
confirmation bias
tendency to look only at info that agrees with what we already believe
bias blind spot
the belief that we are uniquely to fall prey to the other biases previously described
reliability
how consistent the results of a measure are
validity
concerns the operationaliztion measuring what is it supposed to measure
internal reliability
a study participant gives a consistent pattern of answers, no matter how the researcher has phrased the question
criterion validity
evaluates whether the measure under consideration is associated with a concrete behavioural outcome that it should be associated with, according to conceptual definition
known groups paradigm
in which researchers see whether scores on the measure can discriminate among two or more groups whose behaviour is already confirmed
convergent validity
the pattern of correlations with measures of theoretically similar or dissimilar constructs
discriminant validity
something should not correlate with measures of construct that are very different
independent trials
if the outcome of any one trial is not affected by the outcome of any other trial
subjective definition of probabilit
the probability of an outcome is defined by personal probability, your degree of belief that the outcome will occur based on available information
sample space
for a random phenomenon, the sample space is the set of all possible outcomes
event
a subset of the sample space
good story bias
people tend to believe convincing stories
deduction
the process of formulating a prediction that follows from your theory
demand characteristics
a participant wants to be a ‘good’ participant
self perception
people do not necessarily have a correct self image
social desireability
participants want to give a good first impression about themselves
primacy effect
the effect of being the first to be observed
recency effects
the effect of being the last to be observed
test retest reliability
strength of an association between test and retest gives an indication of reliability
cronbachs alpha
a measure of internal reliability
external validity
if it can be generalized
covariance
do the two variables go together
temporal precedence
did the cause occur before the effect
cluster sampling
sample random schools, use children in each school
multistage sampling
a random sample of clusters, then a random sample of people within those clusters
stratified random sampling
in which the researcher purposefully selects particular demographic categories or strata, and then randomly selects individuals within each of the categories, proportionate to their assumed membership of the population
oversampling
blood type ab only 0,5 of population, sample extra ab people to precent unreliability
systematic sampling
each fifth person in a class
convenience sampling
a sample that is easily accessible
purposive sampling
search for participants that meet requirements
snowball sampling
start with a couple of subjects, and ask these subjects whether they know more subjects
quota sampling
set a quota (50 law 50 psych students) and select non randomly up until quotas are fulfilled
observer bias
observations are influenced by your expectations
reactivity
people may behave differently when they know that they are being observed
spurious association
association due to a third variable
posttest design
random assignment multiple groups, test the difference after group assignment
repeated measures design
most important within group design, measure twice in each participant, test the difference
discrete variables
not all variables have the same probability
standard deviation
indicates the amount of dispersion in the population
sampling distribution
of a statistic is the probability distribution that specifies probabilities for the possible values the statistic can take
central limit theorem
for any large enough sample, the sampling distribution is approximately normally distributed
confidence interval
interval containing the most believable values for a parameter
confidence level
the probability that this method produces an interval that contains the parameter
type 1 error
false positive, stating that there is an association when there is no association
type 2 error
stating that there is no association when in fact there is one
internal validity
is an indication of a study’s ability to eliminate alternate explanations for the association
semantic differential format
instead of degree of agreement, respondent might be asked to rate a target object using a numeric scleras that is anchored with adjectives
double-barrelled question
a question that asks two questions in one
response sets
type of shortcut respondent can take when answering survey questions.
acquiescence
yeah saying
fence sitting
playing it safe by answering in the middle of the scale
socially desirable responding
respondents give answers that make them look better than they really are.
observer bias
occurs when observers expectations influence their interpretation of the participants behaviours or the outcome of the study
observer effects
observers change the behaviour of those they are observing
census
if you have the scores of everyone in a population
biased sample
unrepresentative sample, some members of the population of interest have a much higher probability of being included in the sample compared to other members
unbiased sample
representative sample
self-selection
when a sample contains only people who volunteer to participate
probability sampling
every member of interest has an equal and known chance of being selected for the sample, regardless of whether they are convenient or motivated to volunteer.
nonprobability sampling
involve nonrandom sampling and result in a biased sample
weighting
if they determine that the final sample contains fewer members of a subgroup than it should, they adjust the data so responses from members of underrepresented categories count more, and overrepresented members count less.
purposive sampling
if researchers want to study only certain kinds of people, they recruit only those particular participants
confounds
alternative explanations, potential threats to internal validity
design confound
experimenters mistake in designing the independent variable
selection effects
when the kinds of participants in one level of the independent variable are systematically different from those in the other
independent groups design
in which different groups of participants are placed into different levels of the independent variable (between subjects design/ between groups design)
within groups design
(within subjects design) there is only one group of participants and each person is presented with all levels of the independent variable
post-test only design
participants are randomly assigned to independent variable groups and are tested on the dependent variable once.
pretest posttest design
participants are randomly assigned to at least two different groups and are tested on the key dependent variable twice
repeated measures design
a type of within groups design in which participants are measured on a dependent variable more than once, after exposure to each level of the independent variable
concurrent-measures design
participants are exposed to all the levels of an independent variable at roughly the same time, and a single attitudinal or behavioural preference is the dependent variable
power
refers to the probability that a study will show a statistically significant result when an independent variable truly has an effect in the population
order effects
happens when exposure to one level of the independent variable influences responses to the next level
practice effects
in which a long sequence might lead participants to get better at the task or to get tired or bored
carryover effects
in which some form of contamination carries over from one condition to the next
counterbalancing
they present the levels of the independent variable to participants in different sequences
full counterbalancing
in which all possible condition orders are represented
partial counterbalancing
in which only some of the possible condition orders are represented
latin square
a formal system to ensure that every condition appears in each position at least once
demand characteristic
a cue that can lead participants to guess an experiments hypothesis
manipulation check
an extra dependent variable the researchers can insert into an experiment to convince them that tier experimental manipulation worked
interaction effects
the effect of one variable depends on another variable
maturation threat
patient improvement without a specific cause, sometimes called spontaneous remission, can be prevented by : comparison group and pretest post-test
history threat
an external factor that influences the dependent measure, can be prevented by comparison group
regression to the mean
participants with extreme scores will have less extreme results later
non-specific effects
any other effect that causes a change in condition that is unrelated to the treatment
scientific hypothesis
hypothesis about how the world works
statistical hypothesis
hypothesis about a population parameter
p-value
probability of these or extremer results if nulhypothesis is true
level of significance
you conclude that the nulhypothesis should be rejected in favour of the alternative hypothesis. commonly smaller than 0.05 means that the nulhypothesis is rejected
attrition threat
when a certain kind of participant drops out , to prevent this : remove scores from the participant from the pretest too
testing threat
refers to a change in the participants as a result of taking a test (dependent measure) more than once
instrumentation threat
when a measuring instrument changes over time
selection-history threat
an outside event or factor affects only those at one level of the independent variable
selection attrition threat
only one of the experimental groups experiences attrition
null effect
if the independent variable did not make a difference in the dependent variable -> there is no significant covariance between the two
ceiling effect
all the scores are squeezed together at the high end
floor effect
all the scores cluster at the low end
manipulation check
separate dependent variable that experimenters include in a study , to make sure manipulation worked
measurement error
a reason for high within group variability, its a human or instrument factor that can inflate or deflate a persons true score on the dependent variable
weighted average
used when each x value is not equally likely
factorial design
one in which there are two or more independent variables
participant variable
a variable whose levels are selected (measured), not manipulated
marginal means
are the arithmetic means for each level of an independent variable averaging over levels of the other independent variable
file drawer problem
researchers tend to not write papers about insignificant results
publication bias
journals publish insignificant results less often than significant results
data dredging
doing as many statistical tests up until you find a significant resul
harking
hypothesising after the results are known
explorative
no specific hypothesis
confirmatory
you test a specific hypothesis
cross validation
after exploration, conduct a new confirmatory study
stable baseline design
study in which a practitioner or researcher observes behaviour for an extended baseline period before beginning a treatment or other intervention, multiple pretests and posttests, helps to prevent regression to the mean, maturation and nonspecific effect
multiple baseline design
goal is to exclude effect of external factors
direct replication
exactly the same experimental conditions/operationalizations
systematic repication
change some aspects of the study
conceptual replicatio
the same research question, but a complete different experimental setting
principle of beneficence
researchers must take precautions to protect participants from harm and to ensure their well being
bivariate correlation
association that involves exactly two variables
construct validity
how well was each variable measured
statistical validity
how well do the data support the conclusion
effect size
describes the strength of a relationship between two or more variables
statistical significance
refers to the conclusion a researcher reaches regarding the likelihood of getting a correlation of that size just by chance, assuming there is no correlation in the real world.
p value means
the probability that the samples association came from a population in which the association is zero
restriction of range
if there is not a full range of scores on one of the variables in the association it can make the correlation appear smaller than it really is.
curvilinear association
in which the relationship between two variables is not a straight line, it might be positive up to a point and then become negative
directionality problem
sometimes called temporal precedence, when you don’t know which variable comes first
spurious association
the bivariate correlation is there but only because of some third variable.
moderator
when the relationship between two variables changes depending on the level of another variable
quasi experiment
differs from a true experiment in that the researchers do not have full experimental control
nonequivalent control group design
different participants at each level of the independent variable. it has at least one treatment group and one comparison group, but participants have not been randomly assigned
nonequivalent control group pretest/posttest design
the participants were not randomly assigned to groups, and were tested both before and after some intervention.
interrupted time-series design
a quasi experimental study that measures participants repeatedly on a dependent variable before during and after the ‘interruption’ caused by some event.
nonequivalent control group interrupted time series design
it combines non equivalent control group design dnt he interrupted time series design
multiple baseline design
researchers stagger their introduction of an intervention across a variety of individuals, times or situations to rule out alternative explanations
reversal design
a researcher observes a problem behaviour both with and without treatment but takes the treatment away for a while to see whether the problem behaviour returns
which non parametric statistic is appropriate for a research design that compares a quantitative response variable for two independent groups?
wilcoxon test
which non parametric statistic is appropriate for a research design that compares a quantitative response variable for two independent groups?
wilcoxon test
interquartile range
is the distance between the third and first quartiles
interquartile range
is the distance between the third and first quartiles
direct replication
researchers repeat an original study as closely as they can to see whether the effect is the same in the newly collected data.
conceptual replication
researchers explore the same research question but use different procedures
replication plus extension
researchers replicate their original experiment and add variables to test additional questions
p hacking
the goal is to find a p value of just under 0.05, the traditional blue for significance testing and researchers may try to make their results work
preregistration
scientists can preregister their study’s method, hypotheses or statistical analyses online, in advance of data collection
preregistration
scientists can preregister their study’s method, hypotheses or statistical analyses online, in advance of data collection
ecological validity
study’s similarity to real world contexts
theory testing mode
researchers are then designing correlational or experimental research to investigate support for a theory
generalization mode
when researchers want to generalise the findings from the sample in a previous study to a larger population
experimental realism
lab experiments create situations in which people experience authentic emotions, motivations and behaviours
assumptions for a confidence interval that have to be satisfied
random sample and distribution has to be normal
specifity
says negative when it is indeed negative
sensitivity
says positive when it is indeed positive
model
a simple approximation for how variables relate in a population
power depends on ;
size of the effect that you want to detect, dispersion in the population, level of confidence, sample size
