Powerpoint, Week 3 (Quantitative Research) Flashcards
(37 cards)
_____ research is 1) objective 2) deductive 3) predetermined structure (lots of procedures/methods)
Quantitative
______ research is 1) subjective 2) inductive 3) process-oriented structure
Qualitative
Which type of research moves from general -> specific? Inductive or Deductive
Deductive
Which type of research moves from looking at specific cases -> generalize (about that person/not necessarily about population)
Inductive
Research = correlation of relationships. True or False
True (‘bottom of quantitative research totem pole’)
Experimental research design is also known as the “ “?
Gold Standard
Three concepts needed for experimental research (MCR):
Manipulation (changing variables)
Control (researcher has control over experiment)
Randomization (assignment & sampling are 2 kinds, in order to make outcomes freer of biases)
There are 5 different types of control that researchers can exert: (here is the first letter of each kind): A\_\_\_\_\_ \_\_\_\_\_\_ P\_\_\_\_\_\_ S\_\_\_\_\_\_\_ \_\_\_\_\_ \_\_ \_\_\_\_\_ D\_\_\_\_\_\_ \_\_\_\_/\_\_\_\_\_\_ W\_\_\_\_
Alternative Intervention Placebo Standard Methods of Care Different Doses/Intensities Wait-list
Name this experimental design (FYI, when you see R for randomization, you know it’s an experimental design; X is treatment; O1 is posttest)
R X O1
R O1
Basic posttest-only design
- usually done in pilots
- outcome is not relevant until the intervention is complete
- drawback: doesn’t allow an evaluation of whether the 2 groups were comparable at the start (because there was no pretest)
Name this experimental design (FYI, O1 is pretest, O2 is posttest, O3 another repeated follow-up posttest)
R O1 X O2 O3
R O1 X O2 O3
Basic pretest-posttest design (with optional repeated follow-ups)
- want to find out baseline data, data after the treatment is administered, and then again a couple months later
- this is good when the focus of the intervention is on change and when the researcher wants to assess both group differences (experimental comparison) and change within groups (quasi-experimental)
- drawback - sometimes the pretest itself can affect the outcomes of interest
Name this experimental design (FYI, Xa and Xb are two different treatments)
R O1 Xa O2
R O1 Xb O2
R O1 O2
Multiple Intervention Design
- looks to see if there is change with the 2 treatments plus change with the control
- this type of design is good to disentangle the effects of different components of a complex intervention (or to test competing interventions)
- this does require a larger sample size and may be at risk to statistical conclusion validity if a and b are not very different
Name this experimental design
R O1 X O2 O3
R O1 O2 X O3
Wait-list or delay of treatment design (you can see that the first group has the treatment administered while the second group is acting as the control…then a posttest occurs and the second group gets the delayed treatment, while the first group doesn’t get any, since they received it the first round)
- good for when there is patient preference for the innovative treatment and it can strengthen inferences by virtue of replication aspect for the second group
- drawbacks are that the controls may drop out before they actually receive the delayed treatment and it’s not suitable if key outcomes are measured long after treatment (i.e. mortality) or if there is an interest in assessing long-term effects (wait-list period is then too long)
Name this experimental design
R O1 Xa O2 Xb O3
R O1 Xb O2 Xa O3
Crossover design (participants are serving as their own controls)
- good for when recruitment is difficult because you can have smaller sample sizes here (excellent for controlling confounding variables)…it’s also good for when there is no expectation of carryover effects from one period to the next (effects should have rapid onset, short half life)
- drawbacks are that you cannot assume there are no carryover effects and if the first treatment “fixes” the problem, the participants may not remain in the study for the second treatment; also history threat to validity is a possibility
True or False - there are 4 phases to clinical trials
True
Which phase of a clinical trial does this describe (1, 2, 3, or 4)
- initial development of drug or therapy
- designed for safety and tolerance/optimal dose (outcome is looking to see if people can tolerate the drug)
- small scale studies using simple designs
Phase 1
Which phase of a clinical trial does this describe (1, 2, 3, or 4)
- preliminary evidence of treatment effectiveness
- assess feasibility of launching rigorous test
- PILOT test of treatment
Phase 2
Which phase of a clinical trial does this describe (1, 2, 3, or 4)
- FULL test of treatment
- RCT standard design used at this phase
- develop evidence about treatment efficacy (validates effectiveness/efficiency)
Phase 3
Which phase of a clinical trial does this describe (1, 2, 3, or 4)
- effectiveness of intervention in general population
- focus on postapproval safety surveillance
- focus on long-term consequences over larger population
Phase 4
These are some weaknesses of an experimental design:
- strongest evidence for intervention effects
- least bias
- causal relationships
False, these are STRENGTHS
These are some weaknesses of an experimental design:
- impractical
- limited control in clinical settings
- Hawthorne effect
True (nobody does RCT in clinics because it is difficult to randomize in a clinic)
This word describes the knowledge that being in a study can affect a participant’s behavior
Hawthorne effect
Are RCT or pilot studies more often done in clinics?
pilot studies
Name this quasi-experimental design
O1 X O2
O1 O2
Nonequivalent control group, pretets-posttest design
- there is no randomization here
- this is good when an entire unit must get the intervention and a similar unit not getting the intervention is available
- a drawback is that selection threat occurs here (when there’s no randomization, the two groups are different and the outcomes could be attributed to the group differences vs. the IV)
Name this quasi-experimental design
X O1
O1
Nonequivalent control group, posttest only design
- this is when researchers already have some prior knowledge about the comparability with regard to key outcomes
- drawbacks: vulnerable to selection threat as well as history threat
