unit three: research methods (baron's two, mod three) Flashcards
hindsight bias
people have the tendency upon hearing about research findings (and many other things) to think the they knew it all along
ex: hearing a study about how babies sync their heart beat with their mothers and stating you already knew that but you don’t because you didn’t do the research to prove your claim
applied research
comparison of two different methods etc.
this type of research is clear and has practical applications
ex: comparing two different methods of teaching children to see which is more beneficial
basic research
explores the questions that are of interest to psychologists but are not intended to have immediate, real world applications
ex: studying how people form their attitudes about others and how people in different cultures define intelligence
hypothesis
expresses a relationship between two variables
ex: religion, stress level and height are variables.
dependent variable
depends on the independent variable. a change in the independent variable would also alter the dependent variable.
ex: # of hours of playing violent games impact on children’s behaviors
independent variable
the variable that alters the dependent variables result.
ex: amount of fruits eaten vs health of child
theory
aims to explain some phenomenon and allows researchers to generate testable hypothesis with the hope of collecting data that support the theory.
ex: researcher a. believes that if a child eats more fruits they are more likely to be healthy than a child who does not
operational definitions
when you operationalize a variable, you explain how you will measure it.
raises many issues about the validity and reliability of the research.
ex: in the hypothesis used before: what programs will be considered violent? what behaviors are considered aggressive?
research is valid…
… when it measures what the researcher set out to measure
research is reliable…
… when it can be replicated; it is consistent.
participants
the individuals on which the research will be conducted on
sampling
the process by which participants are selected
population
includes anyone or anything that could possibly be selected to be in the sample
sample
the group of participants
representative
the goal in selecting a sample is that it is representative of the larger population
random selection
every member of the population has an equal chance of being selected
true random selection
choosing people by standing in front of a library on a wednesday morning is NOT random because a specific group of people in the population usually goes to the library wendesday mornings (senior citizens)
random selection is best done by a computer, a table of random numbers or by simply picking names out of a hat
stratified sampling
a process that allows a researcher to ensure that the sample represents the population on some criteria.
laboratory experiments
conducted in a lab highly controlled environment,
field experiments
conducted out in the world
experiments
psychologists preferred method of research because only through a carefully controlled experiment can one show a casual relationship.
experiments allow the researcher to manipulate the independent variable and control for confounding variables
confounding variables
any difference between the experimental and control conditions which can affect the dependent variable.
ex: in order to show that the violent games cause aggression I need to rule out that there are not any other reasons for aggression.
assignment
the process by which participants are put into a group, experiment or control.
random assignment
means that each participant has an equal chance of being placed into any group. this prevents participant-relevant confounding variables.
participant-relevant confounding variable
if participants were given the opportunity to choose whether to be in the group watching the violent television or not, it is highly unlikely that the two groups would be compromised of similar people
controlling for participant-relevant variables
using random assignment diminishes the chance that participants in the two groups differ in any meaningful way.
group matching
if one wanted to ensure that the experiment and control group were equivalent on some criterion
ex: sex, IQ scores, age
situation-relevant confounding variables
can also affect an experiment
environment of participants must be similar
making the environments of the two groups as similar as possible controls for situation-relevant confounding variables
ex: if one group watches a video in a lecture hall vs.. another in a classroom then their situations are not equivalent
experimenter bias
special kind of situation-relevant confounding variable.
the unconscious tendency for researchers to treat members of experimental and control groups differently to increase the chance if confirming their hypothesis. not conscious (that’d be fraud)
double-blind procedure
occurs when neither the participants nor the researcher are able to affect the outcome of the research.
can be accomplished by: the researcher to have someone blind to the participants condition interact with the participants
single-blind procedure
occurs when only the participants do not know to which group they have been assigned; this strategy minimizes the effect of demand characteristic (cues about the study) as well as certain kinds of response or participant bias.
demand characteristics
cues about the purpose of the study. participants use such cues to try to respond appropriately. response or subject bias is the tendency for subjects to behave in certain ways.
social desirability
a kind of response bias, the tendency to try to give answers that reflect well upon them
ex: phone polls and people who were afraid to say they would vote trump even though they were anon
experimental group
(IV) the one that gets the treatment operationalized in the independent variable.
control group
(placebo) the control group gets none of the independent variable. it serves as a basis for comparison. without a control group, knowing whether changes occurred in the experimental group are due to the experimental treatment is impossible essentially.
the hawthorne effect
merely selecting a group of people on whom to experiment has been determined to affect the performance of that group, regardless of what is done to those individuals.
placebo method
whenever participants in the experiment groups are supposed to ingest a drug, participants in the CONTROL group are given an inert but otherwise identical substance. researchers can separate the psychological effects of the drug from the psychological effects of those witnessing the placebo effect.
placebo effect
psychological effect of people thinking they took a drug that works even though its just a faux drug they believe it did something because the were either told it was or just inferred it would.
counterbalancing
sometimes using participants as their own control group is possible through counterbalancing.
ex: if i wanted to see how frustration affected performance on an IQ test, I could have my participants engage in a task unlikely to cause frustration, test their IQ then give them a frustrating task and test their IQs again. this causes order effects though.
order effects
participants may do better on the second IQ test simply by virtue of having taken the first IQ test. this can be eliminated by counterbalancing.
counterbalancing
by having half the participants do the frustrating task first and half the participants do the not-frustrating task first then switching.
correlation
expresses a relationship between two variables without ascribing cause. can be either positive or negative.
positive correlation
between two things means that presence of one thing predicts the presence of the other.
negative correlation
means that the presence of one thing predicts the absence of other.
ex post facto
after the fact study.
ex: my hypothesis is that boys are more likely to call out in class. i can’t randomly assign subjects to conditions. the IV has been predetermined. i won’t be able to isolate the cause of calling out behavior. could be biological or societal influences that acted on the two sexes from birth onward.
if i seek to control all other aspects of the research process as I would in an experiment then I will have conducted an ex post facto
survey method
involves asking people to fill out surveys. can not show a cause and effect relationship.
naturalistic observation
researchers opt to observe their participants in their natural habitats without interacting with them at all. the goal of this is to get a realistic and rich picture of the participants’ behavior. control is sacrificed.
case study method
is used to get full, detailed picture of one participant or small group of participants. cannot be generalized to a larger population
ex: clinical psychologists often use case studies to present information about a person suffering from a particular disorder. case studies allow researcher s to get a good picture on what they are studying but is limited to an individual or small groups.
descriptive statistics
simply describes a set of data.
frequency distribution
ex: if i wanted to know what kinds of pets my peers have i’d make a frequency distribution showing how many of them have dogs, cats, bird etc.
can be turned into line graphs called histograms/frequency polygons/ bar graphs
frequency polygons/ bar graphs known as histograms
the y-axis always represents the frequency
the x-axis is whatever else is being graphed (ie: dogs cats, etc.)
central tendency
measures of central tendency attempt to mar the center off a distribution. three common measures of central tendency are the three Ms: mean, median and mode.
mean
the average of all the scores in the distribution.
add up all the scores in the distribution then divide by the # of scores.
median
the central score in a distribution.
write the numbers in order numerically then find the number in the center.
mode
the score that appears most frequently.
distribution can have multiple modes.
bimodal, trimodal, etc.
extreme scores or outliers
ex: 19/20 of my friends have cars all valued at $12,000 but 1/20 has a ferrari valued at $120,000 this skews the data to say that the mean value of our cars is $17,400
unless a distribution is symmetrical it is skewed.
positively skewed
distribution includes an extreme score or group of scores that is very high
…
….. ….
….. ………….
negatively skewed
when the skew is caused by particularly low score (or group of scores) the distribution is negatively skewed.
….
……
.. ……….
measures of variability
another type of descriptive statistical measures. again, you may be familiar with some of these measures: range, variance, and standard deviation.
range
the distance between the highest and lowest score in a distribution
standard deviation
related to the variance. the standard deviation is the square root of the variance.
z scores
the distance of a score from the mean in units of standard deviation.
scores below the median are negative z scores.
ex: Clarence got a 72 on a test with a mean of 80 and standard deviation of 8 his z score is -1.
normal curve
theoretical bell shaped curve for which the area under the curve lying between any two z-scores has been predetermined.
percentile
indicate the distance of a score from 0. someone who scores in the 90th percentile on a test has scored better than 90 percent of the people who took the test too.
correlation coefficient
range from -1 to +1. -1 is a perfect negative correlation and +1 is a perfect positive correlation. both -1 and +1 denote equally strong correlations and 0 is the weakest possible being that it is no correlation
scatter plot
a scatter plot graphs pairs of values, one on the y-axis and one on the x-axis. the closer the points come to falling on a straight line, the stronger the correlation is.
line of best fit or regression line
line drawn through the scatter plot that minimizes the distance of all points from the line. when the line is upwards left to right it indicates a positive correlation.
inferential statistics
used to determine whether or not findings can be applied to the larger population from which the sample was selected.
sampling error
guaranteeing that a sample is representative of a population is impossible.
the extent to which the sample differs from the population is known as sampling error
p value
the probability that the difference between the groups is due to chance.
scientists have decided that a p-value of 0.5 is the cutoff for statistically significant results
statistically significant
0.5 : a 5% chance exists that the result occurred by chance. p value can never be 0 because we can never be 100% certain that results did not happen due to chance.
international review board (IRB)
any type of academic research must first propose the study to the ethics board or the IRB.
they review research proposals for ethical violations or procedural errors. they give the go ahead with the research or requires the revise the procedures.
animal research
IRB requires
- must have clear scientific purpose
- must care for animals humanely
- must acquire animals legally
- design experiment for least suffering possible
human research
IRB requires:
- no coercion must be voluntary
- informed consent
- anonimosity and confedentiality: must not identify source of data
- risk must be minimal
- debriefing
representative sampling
sampling in a way (population) that accurately reflects the characteristics of the population as a whole
illusory correlation
when a person believes that a relationship exists between two variables when it does not
skew right vs skew left graph
RIGHT/POSITIVE
- head to left tail to right
- positive whale cuz it’s going west to cali
LEFT/NEGATIVE
- head to right tail to left
- negative cuz it’s going east to NY
