Inferential Stats/Exp Research (wk 6) Flashcards
Which group (population/sample) contains all subjects of interest?
Population group
Which group (population/sample) gives you the quantity from it, also known as a statistic?
Sample group
Which group (population/sample) is the best approach available?
Sample group
Which group (population/sample) gives you the numerical value of a parameter?
Population group
Which group (population/sample) is a part of selection that is required to be random?
Sample group
Which group (population/sample) is impractical to test every member?
Population group
T/F A sample group does not have to be a good representative of the population group
False, it DOES
Why may a sample not be good even when a truly random sample is used?
- random selection implies that all members of a population have an equal opportunity chosen
- random selection doesn’t guarantee proportional representation of all parts of the population
What is the differences between the sample values and the population values?
Sampling errors
What is the distribution of values taken by the statistic in all possible samples of the same size from the same population?
Sampling distribution of a statistic
What is the distribution of sample means?
Sampling distribution of means (s/square root (n))
where s = sample SD
n = sample size
T/F the sampling distribution of means will be in the shape of the normal curve by Central Limit Theorem
True
Ex: Sampling distribution of means sample of n = 10 mean = 115 sem= 30/square root (10) = 9.5 Interpret this using CI = 95%
95% of the time, the mean might be in the interval
115-29.5, 115+29.5) = (96,134
What is a common inferential stat used to estimate population parameters?
CI
ex: Determine length of hospital stay for patients with LBP from the sampled medical records
WHat is a common inferential stat used to compare between groups?
SHT or CI
ex: Is one treatment more effective than another?
What value will quantify how consistent your sample values are with the null hypothesis?
P-value
a large p-value closer to 1 -
your sample values are consistent with the null hypothesis
a small p-value closer to 0 -
your sample values are not consistent with the null hypothesis
alpha = .05, p-value
leads you to reject the null hypothesis - observed difference shows significant effect
alpha = .05, p-value >.05 leads you to do what?
leads you to retain the null hypothesis - observed difference is probably due to chance and is not significant
What is a range of scores with specific boundaries, or confidence limits?
CI
CI can be used to do what two things?
- estimate population parameters : lumbar spinal extension for 30-39 year old individuals is from 37.3 degrees to 42.7 degrees
- compare between groups:
- the mean difference of SBP between the group with the new medication and the group with the standard treatment was 10 and its 95% CI was (5.1, 15.7)
- the odds ratio of having breast cancer for the group physically active over the group not physically active was 2 and its 95% CI was (0.5, 4.5)
The narrower CI implies ____ precision and ___ variability
higher precision and less variability
The wider CI implies ____ precision and ___ variability
lower precision and increased variability
The null hypothesis: mean of experimental group is the same as the mean of the control group. So mean difference will be a value of what?
mean difference = mean SBPexperimental - mean SBPcontrol = 0
The null: odds of of something is the same for the cases and the controls. So odds ratio will be a value of what?
odds ratio = “oddscases” /”oddscontrols” = 1
if the CI contains the null value, what does that mean?
Effect no significant
if the CI doesn’t contain the null value, what does that mean?
Effect is significant
What term is often used to describe experimental studies that examine the effect of intervention on human subjects?
Clinical Trial
The phases of trials are intended to provide different types of information about what 3 things?
- the treatment in relation to dosage
- safety and efficacy
- with increasingly greater rigor in demonstrating the intervention’s effectiveness and safety
Which trial is to show a new therapy is safe, data collected on dosage, timing, and side effects, typically done on small samples of 20-80 subjects?
Phase 1 trial
Which trial is to demonstrate the effectiveness of a therapy, done on small samples and can take up to 2 years?
Phase 2 trial
The response rate in a phase 2 trial needs to be what compared to standard treatment in order to continue testing?
Same or better
What trial is a randomized and blinded experiment that compares new therapy with standard treatment for placebo, larger scale studies?
Phase 3 trial
If phase 3 trial is successful, what occurs next?
Seek approval from FDA
Describe phase 4 trial stage of a new therapy
Once it has been approved, research continue to investigate effects in other populations to learn about risk factors, benefits, optimal use patterns
What design includes subjects that are randomly assigned to at least two comparison groups?
True-experimental Design
What design does not meet requirements of true experiment, lacks random assignment or comparison groups, but is used when true=experimental design is not feasible?
Quasi-experimental Design
What type of design is when subjects are randomly assigned to independent groups?
Between-subjects design
What type of design is when subjects act as their own control (repeated measures design)?
Within-subjects design
What is a single-factor design (1 -way design)?
- structures the investigation of one independent variable
2. easy to develop
What is a multi-factor design (= factorial design; multi-way design) ?
incorporates two or more independent variables
ex: 2x3 design with two factors Gender (M, F) and BMI level (low, medium, High)
What is a between subject pretest-posttest control group design?
- compares two or more groups that are formed by random assignment
- changes from pretest to posttest that appear in the experimental group but not the control group can be reasonably attributed to the intervention
- considered the scientific standard in clinical research for establishing a cause-and-effect relationship
What is a between subject posttest-only control group design? When would this strategy be used?
- identical to the pretest-posttest control group design but with no pretest administered to either group
- used when a pretest is either impractical or potentially reactive
What is a within-subjects repeated measures design?
- uses subjects as their own control
- potential for practice effects (or learning effect)
- potential for carryover effects (long enough wash out period should be allotted)
When would a within-subjects repeated measures design be used?
only used when the outcome measure will revert back to baseline between interventions and when the patient problem will remain relatively stable through the study period
What is a within-subjects crossover design?
half the subjects receive treatment A followed by B; half receive B followed by A
When would a within-subjects crossover design be used?
- only used when the patient’s condition will not change over time
- not a reasonable approach when treatment effects are slow
- washout period is essential
What is a mixed design?
has both the within-subjects factors and between-subjects factors
