Research Design exam 2

questionnaire research

• series of questions for the purpose of gathering information from respondents

population

all members of a particular group

random sampling

• a group of observations drawn from a larger population

random sampling

• procedure that ensures every person in a population has an equal chance of being chosen to participate

sampling

• researchers make inferences about a population from a sample

internal validity

• refers to your confidence that the change in the DV was caused solely by the IV and not extraneous variables

• how sure are we that IV caused DV

threats to internal validity: history

• any event, other than the IV, that occurs between repeated measurements

threats to internal validity: maturation

• experimental changes that occur over time

threats to internal validity: testing

• when pretesting contaminates later performance

threats to internal validity: instrumentation

• when a method or observer changes what is measured

threats to internal validity: statistical regression

• when selecting patients based on extreme scores leads to regression to the mean on subsequent scores

threats to internal validity: selection

• when groups are not equal before an experiment, biasing the results

threats to internal validity: selection interaction

• when the selection method interacts with one or more of the other threats

threats to internal validity: mortality

• experiment dropout or attrition

external validity

• whether the results of an experiment apply to people and situations that are diferent from those of the experiment

threats to external validity: reactive arrangements

• artificial conditions within the lab do not exist in the world

threats to external validity: sample characteristics

• the composition of your sample can affect generalizability

threats to external validity: convenience sampling

• sampling based on the ease of locating participants

spectrum of research

• observational research

• case studies

• archival research

• questionnaire research

• experimental research

basic research

• goal: develop basic knowledge

applied research

• goal: understand and address problems

qualitative research goal

• understand and describe behavior through non-statistical techniques

quantitative research goal

• explain behavior through statistical inference

observed research

• describe and measure people’s behavior in everyday situations

• patients do not know they are being recorded

case study

• comprehensive description of an individual and his or her behavior

• does not infer cause and effect

• IRB approval

archival research

• systematic examination and inter presenting of existing records in order to answer a research question

• no need for IRB approval

4 main ways of assigning #’s to different levels of a variable

• nominal

• ordinal

• interval

• ratio

categorical variables

• nominal

• ordinal

continuous variables

• interval

• ratio

nominal scale

• assigning any number to take the place of a group or category name

ordinal scale

• assigning numbers to stand for relative rankings

interval scale

• assumes intervals (distance between scores) are equal in size

• arbitrary zero point

ratio scale

• just like interval, but has a meaningful 0

• 0 means the complete absence of the attribute being measured

self report

• unstructured: open-ended

• structured: responses provided

order effects

• the position in a series partly determines the patients responses

carryover effects

• effects that “carry over” from one treatment to another

participant bias: demand characteristics

• cues in a study that could lead patients to be aware of what a researcher is investigating or to respond so as to appease the experimenter

ways to reduce demand characteristics

• use effective cover stories

• administer DVs outside the experimenter context

• avoid self-report measures

• use involving activities

experimenter bias: experimenter expectancies

• beliefs about how patients will or should behave in a study

• researcher gives off subtle but biased cues that elicit the hypothesized responses

ways to reduce experimenter expectancies

• use double-blind procedure

• follow an experimental script

reliability

• consistency of measurement

inter-rater reliability

consistency between or among raters/ observers of some phenomenon

test-retest reliability

• consistency of a measure from one time to another

internal consistency

• consistency

validity

• the extent to which a measure assesses what it purports

construct validity

• the extent to which a measure assess an underlying theoretical construct

face validity

• the extent to which a test seems to measure what it is supposed to measure

convergent validity

• the extent to which a measure correlates with related variables

discriminant validity

• the extent to which a measure does not correlate with unrelated variables

content validity

• the extent to which test items represent the content areas of the underlying theoretical construct

criterion validity

• the extent to which a test or measurement accurately predicts or correlates with an established external outcome or standard

descriptive statistics

• used to summarize and describe a set of data

frequency distributions (histograms)

a graph plotting values of observations on the horizontal axis, with a bar showing how many times each value occurred in the data set

the “normal” distribution

• bell-shaped, symmetrical around the center

skew

• the symmetry of the distribution

positive skew

• scores bunched at low values with the tail pointing to high values

negative skew

• scores bunched at high values with the tail pointing to low values

kurtosis

• refers to the “heaviness”” of the tails

leptokurtic

• positive kurtosis

• heavy or long tails

platykurtic

• negative kurtosis

• light or short tails

central tendency

• measures of central tendency describe the center of a distribution

• mean, median, mode

mean

• arithmetic average value of a distribution

• the sum of scores divided by the number of scores

• M or x̄ (sample)

• µ (population)

median

• the middle score when scores are ordered

• only appropriate for internal/ratio data

when is the median useful

• when you have outliers or skewed data

mode

• the most frequently occurring score in the distribution

bimodal

• two modes

multimodal

• several modes

range

• the smallest score subtracted from the largest

• very biased by outliers

variance

• represents the total amount of similarity or dissimilarity of scores within a sample or distribution

standard deviation

• square root of the variance

correlation

• examines the extend to which two continuous variables are related

• measures the pattern of responses across variables

pearson correlation coefficient

• a statistic that quantifies a linear relationship between two scale variables

p-value

significant value

r-value

pearson correlation