Brainscape's Knowledge GenomeTM


Top Flashcards (Certified)
All Flashcards

Research > Lec 5 > Flashcards

Lec 5 Flashcards

1
Q

parametric statistics are used to analyze ___ data

A

quantitative

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

non-parametric statistics are used to analyze ___ data

A

qualitative

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

when is non-parametric stats used?

A

when assumptions are violated
for ordinal and nominal data

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

T/F: t-tests and ANOVAs are type of regression analyses.

A

T

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

a linear regression shows a significant relationship if the slope is

A

0

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

what are the 5 main parametric assumptions?

A
  1. interval/ratio data
  2. normality
  3. homogeneity of variance
  4. free of extreme outliers
  5. independence of observations
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

how is the normality assumption assessed?

A

check histograms
skewness/kurtosis must be <2 or >-2
Shapiro-Wilk test >0.05

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

what does the Shapiro-Wilk test measure?

A

normality
if there is a difference b/w sample and normal distribution

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

normality is a concern for sample sizes smaller than

A

30
*unless population normally distributed

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

what does homogeneity of variance (HOV) mean?

A

the variance of the outcomes variable should be about the same in each group

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

how is HOV assessed?

A

Levene’s Test

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

what is a good Levene’s test?

A

if the alpha is >0.05
this means that there is no significant difference in variance between groups

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

how are influential outliers assessed?

A

histograms
skewness/kurtosis
boxplots
Cook’s Distance (regression)

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

what is a good outcome for Cook’s Distance?

A

<1 = no influential outliers

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

how is independence of observation assessed?

A

score must not follow a pattern
1 participant’s score can’t influence other

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

if a study uses 1 participant’s involved leg and uninvolved leg as 2 separate groups, what assumption is violated?

A

independence of observation

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

what are the 3 main regression assumptions?

A

linearity
homoscedasticity
outlier testing in regression

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

what is homoscedasticity?

A

the variance of an outcome variable of a relationship/regression test is evenly distributed; it is the same at all levels of the predictor value

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

how is linearity assessed?

A

scatterplot
data points are in a linear pattern

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

what is the residual of a regression test?

A

the distance between the data point and the line of best fit

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

T/F: you want the data to be curvilinear in a regression test.

A

F
want it to be linear

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

how is homoscedasticity assessed?

A

scatterplot
data points should be evenly distrusted around the line of best fit
all residuals should be similar

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

what is the z score equivalent for regression?

A

standardized residuals

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

what are some solutions if an assumption is violated?

A
  1. trim the data (sketchy)
  2. Windsorizing (sketchy)
  3. transform the data (sketchy)
  4. analyze bootstrapping in SPSS
  5. use non-parametric stats
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
if a researcher substitutes outliers with the highest value that isn't an outlier, what is this called?
windsorizing
26
if a researcher deletes a certain number of percentage of score from the extremes, what was done?
trim the data
27
if a researcher using the log value of score to analyze the data, what was done?
transformation of data
28
what is the general goal of hypothesis testing?
rule out chance (sampling error) as a plausible explanation for the results from a research study
29
decision for null hypothesis? the difference between the pre/post samples can be explained by sampling error
fail to reject the null
30
decision for null hypothesis? the difference between the pre/post samples is too large to be explained by sampling error
reject the null hypothesis
31
how is the null hypothesis abbreviated? alternative/experimental hypothesis?
null = H0 alternative = H1
32
which hypothesis states that there is no change in the population before and after an intervention.
null hypothesis
33
which hypothesis states that there is a change in the population following an intervention?
alternative hypothesis
34
the ____ establishes a criterion or cut-off for making a decision about the null hypothesis
alpha level
35
what determines the risk of a type I error?
alpha level
36
what is the critical region (in the tails)?
consists of the outcomes that are vary unlikely to occur if the null hypothesis is true sample means that are almost impossible to obtain if the intervention had no effect
37
how is the test statistic (calculated t) calculated?
difference between the pre & post means divided by the amount of difference one would expect w/o any treatment (std error)
38
is a higher or lower test statistic (calculated t) better?
higher
39
if the test statistic (t) is in the ____, conclude that the difference is significant (reject null)
critical region
40
which is better to be higher calculated t or critical t for positive numbers?
calculate t > critical t negative: cal t < crit t
41
what influences a hypothesis test?
effect size of the intervention sample varibility sample size
42
T/F: a very small effect (change in means) can be significant if it is observed in a very large sample
T
43
what measures the size of the mean difference in terms of the std?
Cohen's d
44
Cohen's d formula
mean difference / std
45
how can the effect size be increased?
increase the mean difference decrease the std
46
what r and d values are a small effect size? % of variance explained?
r = 0.1 d = 0.2 1% of total variance explained
47
what r and d values are a medium effect size? % of variance explained?
r = 0.3 d = 0.5 9% of total variance explained
48
what r and d values are a large effect size? % of variance explained?
r = 0.5 d = 0.8 25% of total variance explained
49
the larger the Cohen's d --> the ___ the effect size and the ____ the power
larger ES and power
50
how can power be increased?
increased effect size - increase mean difference - decrease variability increased sample size increased alpha use a 1-tail test
51
what is power?
the probability that the stat test will reject the null when the treat does have an effect
52
what t test compares data of one group over 2 time periods or twin data?
dependent t test
53
what type of test should be run to determine the amount of participants needed for an independent t test for a good power?
priori power anaylsis
54
T/F: the critical t (critical value) must be reported in the manuscript
F
55
what tests is HOV used for?
independent t tests
56
if the degrees of freedom (ind t test) is ____ HOV is not met .... less likely to reject null
smaller
57
if SPSS reports the p value as 0, how should it be reported in the manuscript?
p ≤ 0.0005
58
which t test is more likely to have a significant result and more power? why?
repeated/dependent t test less sampling error, dof closer
59
which test should be ran? want to determine difference in motivation before and after same group of people watch a film
repeated measures/dependent t test
60
T/F: HOV is an assumption for a repeated t test
F!!
61
what non-parametric stats test can be run on an independent t test?
Mann-Whitney U-test
62
what non-parametric stats test can be run on an dependent/repeated measures t test?
Wilcoxon signed ranks test
63
what value is non-parametric stats run on?
mean rankings
64
what formula is used to determine the true alpha when t tests are repeatedly ran?
1 - (1-a)^# of t tests
65
what is a Bonferroni Correction?
adjusts for alpha level inflation by dividing the alpha by the number of t tests ran

Research (10 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2025 Bold Learning Solutions. Terms and Conditions