quantitative research design Flashcards
cross-sectional design
collect data at only 1 time point
-observational
-measures outcome and exposures in study participants at same time
longitudinal
repeated measurement of variables over time
-observational
-studying same group over time
between groups designs
seperate group for each condition
-participants only provide data once
-> 3 groups 3 different treatment conditions
-observe each once?
repeated-measures designs
-observations are taken from the same participants more than once
-“time of observation” is an independant/predictor variable
mixed factorial designs
both between-group and repeated measures elements
- measuring 2 different groups and different time points
non-experimental designs
research designs where there is a theoretically presumed cause and effect
-no control group and random assignment
-“observational design”, “correlational design”
-> correlational design : relationship bw two variables investigated, without researcher controlling or manipulating any of them
experimental designs
the researcher manipulates the independant variable(s) to determine if it has an effect on the dependant variables(s)
group 1 -> intervention -»»>
group 2 -»»»»»
notation
t = treatment
0 = observations
-> represent when variables are measured, or in other words, when data are collected
R = randomly assigned groups
-> equal probability of being assigned to any of the groups
N = Non-equivalent groups
->groups of participants are typically already intact
pre-experimental designs
participants receive the intervention/treatment of interest
-limited control over threats to internal validity
-participants not randomly assigned to groups
-changes in the dependent variables cannot be attributed to the manipulation of the independant variable
one shot study (pre-experimental design)
can conclude that at the time of the observation participants performed in a certain manner
group A t -> ob
-participants exposed to a intervention and then assesed on the outcome of interest
one-group pretest-posttest study (pre-experimental design)
try to determine the magnitude of the treatment effect
-repeated-measures design
-researcher cannot attribute any changes in performance to the treatment
assessment -> intervention -> assessment
or O -> T -> O
posttest only with non-equivalent groups
static group comparison
-between-groups design
-groups are not randomly formed (intact groups)
-cannot be determined if group A and B are equivalent at the beginning of the study
->differences between 01 and 02 cannot be attributed to the treatment
-one group gets treatment one group does not
N group A T O1
Ngroup B T O2
true experimental designs
may be most powerful means of generating new knowledge
-only quantitative design that can be confidently used to identify cause and effect relationship
-typically conducted in lab in controlled environment -> internal validity
-> less external validity
random assignment (true experimental designs)
equal probability of a participant being placed into a level (group) of the independant variable
-assumes personal factors that could influence participants scores on the dependant variable are distributed evenly across groups (equivalent groups) , thus changes in the dependant variable are most likely due to the manipulation of the independant variable
random assignment helps control for
past history
maturation
testing
random assignment does not control for
-measurement errors
-something happening other than the treatment to one group, but not the other group
-experimental mortality
posttest only control group design
R group A T O1
R group B O2
-theoretical argument that differences between O1 and O2 could be attributed to the treatment
post test only control group design
R group A T1 O1
R group B. T2. O2
R group C O3
-3 levels of the IV
-determine if one treatment is more effective than another treatment and if it is better than nothing
posttest only control group design
-two independant variables
R group A A1B1. O1
R group B A2B1. O2
R group C A3B1. O3
R group D A1B2 O4
R group E A2B2 O5
R group F A3B2 O6
pretest-posttest control group design
R group A O1 T. O2
R. group B O3 O4
-mixed factorial design
-determine how much more change is observed in Group A vs. Group B
-can be extended to include more observations and independant variables
solomon four-group design
R group A O1 T. O2
R. group B O3 O4
R group c T O5
R group D O6
-combines post test only and pretest-post test control group designs
-determination if there is a testing threat to internal validity and if the pretest interacts with the treatment
-replication of treatment effect
quasi-experimental designs
-two or more groups but no random assignment to control and experimental groups
-> uses control groups? textbook
-interested in maximizing external validity
-> want to approximate real world settings
-participants are not randomly assigned
-> not possible
-> participants either self-select themselves to one of the groups or an administrator (eg. coach ) decides who will receive the treatment
ex. post facto design
-groups (IV) are already formed based on a characteristic of the participants and are compared on the DV
-> eg. selected olympic hopefuls vs non selected olympic hopefuls
** no treatment used
-use when IV is not easily manipulated
-want to compare scores on the DV based on already formed groups
-> LEAFS vs Canucks fans
time-series design
01 02 03. 04 T. 05 06 07. 08
-infer cause and effect by establishing the rate of change between observations is different between 04 and 05
-not feasible or practical to have a control group
-> effect of athlete centralization on athletes well being
-participants act as their own controls
control group
group that does not receive the treatment being studied
single-subject design
-effect of the intervention on a single subject
-a repeated measure design
-> time-series
-often used to examine unique/outlier cases
clinical trials
for the purposes of registration, a clinical trial is any research study that prospectively assigns human participants or group of humans to one or more health-related interventions to evaluate the effects on health outcome
eg. drug , exercise program
-measure health related outcome
flow diagram
how many interested in trial
how many met inclusion criteria
how many lossed in trial and how many finished
