Research Design Flashcards
Explain the difference between experimental and non experimental design
Experimental (descriptive, correlational): They are meant to answer questions about causality, the researcher manipulates the variables (treatments) to establish causality (does smoking cause lung cancer?). There are the strongest protection against mistakes.
Non-experimental (experimental, quasiexperimental): They are meant to answer questions that have to do with describing things (what are the characteristics of smokers?) or how characteristics are related (is smoking related to lung cancer?). The researcher does not actively intervene.
An RCT consists of how many levels with one Independent variable what do levels refer to
Consists at least two levels, including:
A treatment group
Some form of comparison or control group
Randomized: Refers to random allocation of participants to
Level is the amount of things the independent variable is comparing like the degrees and test taking situations like you are comparing a 90 degree room to a 50 degree room and seeing which one is better. so it has two levels.
Issues with RCT’s
Is it true random sample???
Since all members of the defined population are available and willing to participate the study.
Some clinical issues…
Clinical procedures are effective or not??? They will work for whom???
RCT’s cannot tell which participants in a sample improved because of treatment and which ones improved because of extraneous factors
Why is randomization good???
Random assignment in experiments is important because:
it helps to control confounding variables.
it ensures that the measures used are valid
it eliminates the problem of dependent variables.
it ensures that the measures used are reliable
Factorial designs and the different levels
think of the room temperature example
Refers to the number of treatments (factors, IV)
One-Way (Single-factor) design: There is only one treatment (factor). It can have more than one level (different dosages, different durations, ..).
Two-way factorial design: Examines two treatments and their interactions.
Three-way factorial design: Examines three treatments and their interactions
what is a single factorial (think only one group with a pre test and post test)
Pretest-posttest control-group design Posttest-only control group design Single-group pretest-posttest design Nonequivalent control group design Time series design Repeated measures or repeated treatment designs
Pretest-posttest control-group design what is this
Think of a placebo group added to a RCt
so you Have treatment group 1 pre o x o post
group 2 o pre D o post and D is the sham or placeboo
group 3 o o o is for a coventional control group
Post-test only
Test group 1 R X O
test group 2 R O
Single group pre and post test design not an RCT or
test group 1 O X O O = testing
X=intervention
Nonequivalent control-group design explain
When a nonrandom control group is available.
It is not RCT but between-groups design
Group 1: O X O
Group 2: O O
Time-Series Design
and advantages
It is used to establish a baseline of measurements before initiation of treatment to either a group or an individual.
If it is important to assess changes over many time intervals
: Group 1: O O O O O X O O O O
to study the effect of something over time
They are widely used in health science research.
Repeated measures for different treatments
Any pretest-posttest designs
Within-subject or within-groups designs
Randomized
explain carry over effect or order effect
think medication wise and how do you eliminate this???
If there are more than two treatments or more than two levels of and independent variable
Order effect: the order in which they are presented may influence the outcome.
Carryover effect: a treatment can enhance or diminish other treatment’s effectiveness
RANDOMIZATION minimizes these effects
Multiple-factor experimental design explain
Researchers interested in:
The individual effects of the multiple factors on the dependent variable
The effects of the interaction between the multiple factors on the dependent variable
explain a 2x2 2x3 3x3
2 x 2 2= groups being used and the 3 represents the number of factors being studied like temp of a 90, 50, room on group 1 and group 2
2X3 same deal just with another degree room like 30 degrees
3x3 3 = another group the first number is the number of groups and third number is the number of factors being tested that affect the dependent variable
explain between groups and within-subject design
then group compared to other individuals
Between-subject design: Each subject is assigned to one group and only receives treatment given to that group. There are separate groups, and comparison is made between groups.
Within-subject design (repeated measures): Each subjects receives all treatments and outcomes for each subject are compared across these treatments