Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

becky: research > LECTURE 2 > Flashcards

LECTURE 2 Flashcards

1
Q

nominal/ordinal is what type of data?

A

qualitative
nominal: disease category
ordinal: assist level

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

IR data is what kind of data?

A

quantitative/parametric stats!
numbers ordered and evenly spaced

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

descriptive stats for quantitative data

A

mean (average)
median (middle of all scores)
mode (most frequent score)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

association question of quantitative data

A

Pearsons correlation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

difference question of quantitative data

A

t-test or ANOVA

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

qualitative data or quantitative data that does not meet assumptions

A

nonparametric stats

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

nonparametric stats descriptive statistics are

A

median and mode

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

nonparametric stats association questions are

A

spearman’s correlation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

nonparametric stats difference questions are

A

wilcoxon-mann whitney
friedman
kruskal-wallis

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

descriptive studies examples

A

JUST DESCRIBES DATA (NO DIFF/RELATIONSHIP QUESTIONS)
-surveys
-retrospective data: previous collected
-normative: normal ppl data collected
-qualitative: analyze interviews to find themes

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

exploratory research study examples

A

RELATIONSHIP BTWN VARIABLES
1. correlational: relationship btwn between outcome measures
2. predictive: can TUG predict LOS?
3. methodological: reliability/validity of outcome measure

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

case control, cohort, and single subject studies are

A

exploratory! looking for differences

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

quasi-experimental studies are

A

cohort (exploratory)
*use two groups with pre-existing variable or pre/post

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

experimental studies are also called

A

RCTs (true experimental design with randomization)
GOAL: demonstrate cause-effect relationship between variables with randomization

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

alpha level of 0.05 is saying what

A

5% risk of committing type 1 error (pre-set significance level)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

p-value is what

A

determined by outcome: actual probability that results were bc of sampling error (should be less than 0.05)

*there is a __% chance that decision to reject the null was wrong!

17
Q

4 types of validity

A
  1. internal
  2. construct
  3. external
  4. statistical conclusion
18
Q

what study is the best design to maximize internal validity?

A

RCT!

19
Q

internal validity could be affected due to….8 things!

A

is Independent variable changed by anything other than dependent variable?
1. history
2. maturation (children, normal recovery)
3. attrition/mortality (dropped out)
4. repeated testing (get better?)
5. instrumentation (did tool change or need calibration?)
6. regression to mean (extreme scores meet towards middle)
7. experimenter bias
8. selection (did groups differ in key ways? did sample not represent pop?)

20
Q

construct validity

A

are we measuring construct we think we are measuring (ex. using ROM for function)

21
Q

external validity

A

can results be generalized to population? (but need to be specific enough to find differences)

22
Q

errors in statistical conclusion validity include…

A
  1. low power (small sample, high variablity)
  2. violation of stat assumptions or used wrong test for type of data
  3. error rate: type 1 or type 2
23
Q

when can you make a type 2 error?

A

when you failed to reject when you should have, usually bc low power or sample size

24
Q

simple random sample

A

everyone in population has equal chance

25
Q

systematic sampling

A

select certain order (every 20th student)

26
Q

stratified sampling

A

sample frame divided into parts: randomly select 10 students from every program

27
Q

cluster sampling

A

sample frame divided into parts but only certain parts are used with all members sampled: randomly select students from 3 randomly selected programs

28
Q

convenience sampling

A

members volunteer or self select (surveys)

29
Q

small effect size (Cohen’s d)

A

0.2

30
Q

medium effect size

A

0.5

31
Q

large effect size

A

0.8

32
Q

how to calculate effect size/Cohen’s D

A

mean1-mean2/ (pooled SD)

becky: research (8 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions