EXAM III Material Flashcards
(101 cards)
1
Q
Define Population
A
All individuals
DIFFERENT from study population (the final group of individuals selected for a study)
2
Q
Define Sample and state when it’s useful to be used
A
A subset or portion of the full population
“representatives”
Useful when studying the complete population is not feasible
3
Q
Study measurements of human studies are collected based on:
A
Desired “variables”
Dependent variable(s) outcome variable
Independent variables
In which comparisons will be made = Statistical analysis
and inferences will be made about the sample-derived measurements and their comparisons (in relation to Null Hypothesis) & to the full population of similar subjects (generalizability)
4
Q
Define Null Hypothesis
A
A research prospective which states that there will be no true difference b/w the groups being compared
Most conservative and commonly utilized
5
Q
What statistical references can be taken by the researcher in a null hypothesis?
A
Superiority
Noninferiority
Equivalency
Don’t always have to show superiority; sometimes equivalency is good enough
6
Q
What are the 2 key attributes of data measurement/variables in which help to determine the statistical test?
A
Magnitude/Dimensionality - i.e. pain level scale, satisfaction, fanciness
Consistency of scale/fixed interval - equal, measurable spacing between units - i.e. date, time, months, age groups, etc.
A 3rd one is rational/absolute zero
7
Q
What are the 3 categories for data/variables based on the answers to the two key attributes of magnitude and consistency of scale?
A
Nominal/Dichotomous/Binary; Non-Ranked named categories
Ordinal/Ranked Categories; equal-distance
Interval/Ratio (order & magnitude & equal intervals-of-scale (units))
8
Q
Define the nominal category for data/variables
A
Has No magnitude, No consistency of scale, No rational zero
Dichotomous, Non-ranked, Categorical
Simply labeled variables without quantitative characteristics
Descrete = Whole Numbers
No mean - cannot interpret mean
i.e. male/female, hair color, eye color, etc.
9
Q
Define ordinal level of data measurement and list examples
A
Yes Magnitude
No Consistency of scale
No Rational Zero
Descrete = whole numbers
May calculate Mean; must be careful tho
i.e. pain level scale, ranking, satisfaction
10
Q
Define Ratio/Interval level of measurement and list examples
A
Yes Magnitude
Yes Consistency of Scale
N/Y Rational Zero (N = Interval, Y = Ratio)
Continous = Fractional Numbers
Can calculate mean, median, standard dev.
i.e. age, number of living siblings, anything physiological measured (BP, lipid panel, etc.)
11
Q
What type of data measurement is applicable while having a Parametric test (normally distributed shape of data distribution)
A
Interval
MMM are essentially equal
Equal dispersion of curve “tails” to both sides of MMM
12
Q
What does it mean to have a positively skewed graph?
A
When the tail points to the right
Mean > Median
13
Q
What does it mean to have a negatively skewed graph?
A
When the tail points to the left
Mean < Median
14
Q
Define skewness, what is the value depicting a symmetric, normal distribution?
A
A measure of the asymmetry of a distribution
Value of 0 = perfectly normal, symmetric, equal MMM
The further from 0, the more the data is skewed
15
Q
Define kurtosis and what (+) and (-) kurtosis means
A
A measure of the extent to which observations cluster around the mean
How well the values cluster around the mean/middle
(+) = more cluster within the graph/bell shape
(-) = less cluster within the graph/bell shape
16
Q
What are the required assumptions of interval data for proper selection of a parametric test?
A
Normally distributed
Equal variable = Levene’s Test; used to determine if the interval data is equal; within the SBSS test
Randomly-derived and Independent
ALL must be true in order to pick Interval measurement
17
Q
What question does Confidence Interval address?
A
What is the plausible range of possible difference or relationship within which I believe the true difference or relationship may lie?
18
Q
What question does the p value address from a statistical test?
A
What is the single measurement value most likely to represent the true (yet unknown) difference or relationship between the groups being compared and what is the probability the difference has occurred by chance?
19
Q
p value is attributed to Type I or Type II error?
A
Type I error = False Positive
When a test detects the presence of disease when in fact the person does not have the disease
20
Q
Define p value
A
The probability of making a Type I error if the Null Hypothesis is rejected.
21
Q
What does is mean when a graph has a kurtosis value of 0?
A
A perfect cluster of obversations at the mean of the bell shaped curve = a normal distribution
22
Q
Which type of data(s) are discrete and why?
A
Nominal and Ordinal
Variables are discrete because they do not have a scale and you cannot have something in between the variables, not continuous
23
Q
What is a Levene’s test and why is it used?
A
A test that is done when you need to determine if the data has magnitude and a balanced scale
Tells us if the data is interval data, if there are equal variances
Asses for equal variances b/w groups
24
Q
What requirements must be met before choosing a Parametric test?
A
Interval data must be:
Normally distributed (bell-shaped curve)
Contain equal variances
Randomly-derived and independent
25
What type of stat test is used for non-parametric tests?
**Descriptive Stats and Graphs**
Data is transformed into a standardized value
**(z-score or log)**
These are stat tests that does not
26
Define Power (1-beta)
The ability of a study design/methodology/selected test statistic to **detect a true difference if one truly exists b/w group-comparisons and the level of accuracy in correctly accepting/rejecting the Null Hypothesis** (analogous to sensitivity in screenings)
Value is traditionally set at **20%**
27
How does the sample size impact the power of stat significance?
The larger the sample size, the greater the likelihood/ability of detecting a difference if one truly exists
**Increase in Power**
**Must add in anticipated drop-outs or loss to follow-ups**
28
In determining sample sizes, would you need a larger and smaller number of samples when the differences between groups are smaller?
Need a greater number (N)
Due to the fact that there is a small difference, it will be harder to find differences in a small group, therefore you need more samples
**Add in anticipated drop-outs or loss to follow-ups**
29
What is another word for a type I error rate?
*p* value
If \< 0.05 it is statistically significant
Determined before a study begins
**Low *p* values** allow more confidence and have a **less risk of Type I error**
by chance alone
30
Define Type I Error and state another term for it
**alpha**
**Rejecting** the Null Hypothesis when it is actually **true** and you **should have accepted it**
There is really NO TRUE DIFFERENCES b/w groups but you in error reject the null hypothesis, thereby stating that you believe there's a difference b/w the groups when in fact, there really isn't!
= **False Positives**
31
Define Type II Error and state another term for it
**Beta**
**Not rejecting** the Null Hypothesis when it is actually **false** and you **should have rejected it**
When there really _IS A TRUE DIFFERENCE_ b/w the groups being compared but you in error do not reject the null and state that there is no difference when there actually is
**= False Negative**
32
What are the most common selected confidence interval values?
90%, **95%**, 99%
33
Explain the interpretation of a 95% CI
We are 95% confident that the "true" difference (0) or relationship (1) between the groups is contained within the CI range
34
List the common types of measures of central tendency
**Mean, Median, Mode**
**Outliers**
**Minimum and maximum range**
**Interquartile Range**
35
Define the two measures of variation/spread/dispersion of data
**Variance** (from the mean) - the difference in each individual measurement value and the groups' mean
**Stardard Deviation -** The square root of variance value
36
What are the 3 population percentages comprised within 1, 2, and 3 SDs around the mean of a normally distributed dataset?
1 = 68%
2 = 95%
3 = 99%
37
Differentiate between dependent and independent data and list the common terms used for dependent data
Dependent data is when you have **data from the same/paired groups = before/after, pre/post, beginning/end, start/finish**
Independent data is when you have **data from different groups**
38
What 4 questions must you ask yourself in while determining a correct stat test to use?
1. What type of data is being collected/evaluated?
**Nominal/Ordinal/Interval**
2. What type of comparison/assessment is desired?
**Correlation, Survival, Regression = stop here**
**Frequencies, Proportions, Counts = proceed to Qs 3 and 4**
3. How many groups are being compared?
4. Is the data independent or dependent?
39
What is a Correlation test, when would you want to use it, what do they tell you and whatare the types of correlation tests for the 3 data types?
Provides a quantitative measure of the **strength and direction**
40
When is a partial correlation test used?
When you are performing a correlation test and want to **control for confounding variables**
41
What is a contingency coefficient test and what type of data is it used for?
A correlational test that determines the strength and relationship between variables
## Footnote
**Nominal Data**
42
What type of test is a Spearman/Kendall Correlation test and what type of data is it useful for?
A Correlational test used to provide a quantitative measure of the strength and direction of a relationship between variables
**Ordinal Data**
43
What type of test is a Pearson Correlation test and what type of data is it useful for? What happens when you have a p-value \> 0.05?
A type of stat test that measures the correlation quantitatively of strength and direction of a relationship between groups
**Interval Data**
If there is a sig. p value, you can say that there is a linear correlation
44
What are survival tests, what are they used for, how can the data be represented as and what are the tests for each of the 3 data types?
Used to measure an event occurrence/proportion of events/time-to-an event
All can be represented as a Kaplan-Meier Curve
Nominal = **Log Rank Test**
Ordinal = **Cox Proportional Hazard Test**
Interval = **Kaplan-Meier Test**
45
What type of test is a log-ranked test and what type of data is it used for?
A type of survival test that compares the occurrence/proportion of events/time-to-an event between groups
**Nominal Data**
46
What type of test is a Cox-Proportional Hazard Test and what type of data is it used for?
A type of Survival test used to compare a proportion of events/occurrence/time-to-an event between groups
**Ordinal Data**
Able to represent data on a Kaplan-Meier Curve
47
What type of stat test is the Kaplan-Meier Test and what type of data is it used for?
A type of survival test used to compare a proportion of event/occurrence/time-to-an-event between groups
**Interval Data**
48
What is a regression test and what are the 3 types of test under each data category?
Tests done to measure the relationship/association between variables by making a prediction about the dependent variable outcome by going back to old data
Able to calculate **OR**
Nominal = **Logistic Regression**
Ordinal = **Multinominal Logistic Regression**
Interval = **Linear Regression**
49
What type of data is used for a logistic regression test?
A regression test that allows you to test the relationships/associations by going back to old data and making a prediction about the dependent variable outcome
**Nominal Data**
50
What type of data is a Multinominal Regression Test used for?
A type of regression test used to measure the relationship/association of groups by going back to old data and making a prediction of the dependent variable/outcome
**Ordinal Data**
51
What type of data is used for a Linear Regression Test?
A type of regression test that measures the relationship/association of groups by going back to old data and making a prediction of the dependent variable/outcome
**Interval Data**
52
What type of tests are used for nominal data when you have 2 groups? (for dependent and independent data)
(before/after, pre/post, start/finish)
Dependent/Paired data = **McNemar Test**
Independent data = **Pearson's Chi-Squared and Fisher's Exact** (able to handle small numbers)
53
What type of tests are used for nominal data when you have 3 or more groups? (for dependent and independent data)
**Independent Data = Chi-square and Fisher's Exact**
**Dependent/Paired Data = Cochran**
All are under the category of **Bonferroni Test of Inequality -** Conservative method of adjusting the p value for the # of comparisons that are being made
54
What is the Bonferroni test of inequality and which tests fall under that category?
**Bonferroni Correction**
Conservative method that adjusts the p value for the number of comparisons being made
55
What is a Pearson's Chi-square test? What type of data is it used for?
A chi-squared test done when you have **Nominal Data with 2 groups that contain independent data**
56
What is a Chi-square test and what type of data is it used for? What must occur if you have a finding that is statistically significant?
A test done with **_Nominal data_ with _Independent data_ of _3+ groups_**
If p-value is stat. significant, must perform a **post-hoc analysis test** to determine which groups are different
More likely to have a type I error if performing multiple chi-squared tests
May perform a **Bonferroni test of inequality** to adjust the p value for the number of comparisons that are being made
57
What is a Fisher's Exact test and what type of data does it use?
Test done using _nominal data_ with _Independent data_ done with **_3+ groups_**
You have an expected cell count of less than 5
This test is able to handle small numbers
58
What is a McNemar test and what type of data does it utilize?
Test done with _nominal data_ when you have _2 Groups_ with **_dependent data._**
dependent/paired = _(_before/after, pre/post, start/finish)
59
What is a Cochran Q test and what type of data does it utilize? What is its relation with the chi-squared test?
Test done with _nominal data_ with _dependent/related data_ done on **_3+ groups_**
If there is a p-value that is stat. significant, must perform a **post-hoc analysis** to determine which group is different
May use a **Bonferroni test of inequality/Bonferroni correction;** a conservative approach that adjusts the p-value for the number of comparisons that are being made
60
What are the 2 tests that are performed with independent variables that contain ordinal data?
**Mann-Whitney test**
**Kruskal-Wallis test - post-hoc test if sig. p value is found**
Both are used for interval data that does not meet the parametric requirements
Both compare the median value between groups
61
What are the 2 tests that are used when you have dependent/paired variables with ordinal data?
**Wilcoxon Signed Rank test**
**Friedman test -** If p-value is sig. must perform a **post-hoc test** to see which group is different
Both tests compare the **median** between groups
Both tests are used with interval data that does not meet parametric requirements
62
What tests are used when you have 2 groups with ordinal data? (independent and paired data)
Independent data = **Mann-Whitney test** (3+ groups, p-value sig. = post-hoc test to find where difference lies)
Dependent/Related data = **Wilcoxon Signed Rank** (pre/post, before/after, start/finish)
Both tests compare the median values
Both tests used for interval data that does not meet parametric requirements
63
What type of tests are used when you have ordinal data with 3 or more groups? (independent and related data)
Independent = **Kruskal-Wallis**
Dependent = **Friedman**
Both tests compare the median values
Both tests must perform a post-hoc test if you find a sig. p-value
Used for interval data that does not meet parametric requirements
64
What are the characteristics of stat tests that require a post-hoc test?
When the test has **3 or more groups** and you have a **statistically significant p-value**
65
What post-hoc tests are used when you have nominal and ordinal data with 3 or more groups? (3)
**Student-Newman-Keul Test -** compares all pairwise comparisons possible; groups must be equal in size
**Dunnett Test -** compares all pairwise comparisons possible against a **single control;** groups must be equal in size
**Dunn Test -** compares all pairwise comparisons possible; **groups do not have to be equal**
66
What is a Dunn test and when is it useful?
A post-hoc test used when you have ordinal data with 3 or more groups (independent OR paired/related)
Compares all pairwise comparisons possible
**Groups do not have to be equal in size;** as opposed to Dunnett and Student-Newman-Keul test
67
What is a Student-Newman-Keul test and what is it useful for?
A post-hoc test done when you have a significant p-value with ordinal data with 3 or more groups and you need to determine where the difference lies.
Compares all pairwise comparisons possible and **groups must be equal in size;** as opposed to the Dunn Test
68
What is a Dunnett test and what is it useful for?
A post-hoc test done when you have a significant p-value and must find where the difference lies when you have 3 or more groups with ordinal data.
Compares all pairwise comparisons against **a single control** and **all groups must be equal in size**
69
What are the possible tests when you have interval data with independent variables? (5)
**Analysis of Variance (ANOVA) -** 3+ groups
**Analysis of Co-Variance (ANCOVA) -** 3+ groups
**Multiple Analysis of Variance (MANOVA) -** 3+ groups
**Multpile Analysis of Co-Variance (MANCOVA) -** 3+ groups
**Student T-test -** 2 groups
70
What is a Student T-test? When is it useful?
Test performed when you have _interval data_ with _2 groups_ with **_independent data_**
Compares the means of all groups, along with intra- and inter-group variations against a **single dependent variable**
71
What is a Analysis of Variance (ANOVA) test and what is it useful for?
Test used when you have _interval data_ when you have _3 or more groups_ with **_independent variables_**
Compares the means of all groups along with the inter- and intra- group variations against a **single dependent variable**
Able to use with 2 groups as well
72
What is a Multiple Analysis of Variance (MANOVA) and what is it useful for?
Test done when you have _interval data_ with _3 or more groups_ that have _independent data_ against **_multiple dependent variables_**
Compares the means of all groups and the inter- and intra- group comparisons
Must use a **post-hoc test** if p-value is stat. sig.
73
What is an Analysis of Co-Variance (ANCOVA) and what is it useful for?
A test done when you have _interval data_ with _3 or more groups_ containing **_independent variables_**
Compares the means of all groups along with inter- and intra- group variations against a **single dependent variable while controlling for the co-variance of confounders**
74
What is a Multiple Analysis of Co-Variance (MANCOVA) test and what is it useful for?
Test used when you have _interval data_ with _3 or more groups_ with **_independent variables_**
Compares the means of groups along with inter- and intra- group comparisons **against multiple dependent variables while controlling for the co-variance of confounders**
75
What tests are useful when you have interval data with 2 groups? (independent and dependent/related)
Independent = **ANOVA, MANOVA, Student t-test**
Dependent/Related = **Repeated Measures ANOVA, Repeated Measures MANOVA, or Paired t-test**
76
What tests are useful when you have interval data with 3 or more groups? (independent and dependent variables)
Independent = **ANOVA or MANOVA**
Dependent/Related = **Repeated measures of ANOVA or Repeated measures of MANOVA**
77
What are the 3 tests that are useful when you have interval data with dependent/related variables?
**Repeated Measures ANOVA**
**Repeated Measures MANOVA**
**Paired t-test**
78
What is a paired t-test and what is it useful for?
A test done when you have _interval data_ with _3 or more groups_ with **_dependent/related data_**
Compares the mean values between groups that are related
79
What is a Repeated Measure ANOVA and what is it useful for?
Repeated Measure Analysis of Variance
Useful when you have _interval data_ with _3 or more groups_ containing **_dependent/related/paired data_**
Compares the means of all groups along with inter- and intra-group variations of **related data** against a **single dependent variable**
_Able to control for confounders_
If confounder is present = **Repeated Measures ANCOVA or Repeated Measures MANCOVA** ---\> if sig. p-value = **Bonferroni test or Scheffe or Dunn or Student-Newman-Keul or Dunnett test**
80
What is a repeated measure MANOVA test and what is it useful for?
Repeated Measure Multiple Analysis of Variance
Useful when you have _interval data_ with _3 or more groups_ with **_dependent/paired/related variables_**
Compares the means of all groups along with inter- and intra- group variations of **related data** against **multiple dependent variables**
Able to control for confounders; however if a confounder is present, must perform a post-hoc test; _Repeated Measures ANCOVA or Repeated Measures MANCOVA_
If p-value is sig. = _Bonferroni test/Tukey/Scheffe/Dunn/Student-Newman-Keul/Dunnett tests_
81
What are the post-hoc tests that are performed when you have interval data with 3 or more groups comparisons?
**Tukey/Scheffe Test**
**Bonferroni Test**
**Dunn/Dunnett Test**
**Student-Newman-Keul Test**
82
What is a Tukey/Scheffe Test? What is it useful for?
A **post hoc test** performed when you have **3 or more groups** with **interval data**
Compares all pairwise comparisons possible
All groups must be equal in size
_Tukey_ - slightly more conservative than Student-Newman-Keul Test
_Scheffe_ - is less affected by violations in normality and homogeneity of variances
83
What is a kappa statistic?
An agreement that is made between evaluators (apart of the validation/assessment committee); consistency of "decisions", "determinations"
+1 = observers classify everyone the same way
0 = no relationship at all between the observers classifications above the agreement that would be expected by chance
-1 = observers classify everyone exactly opposite of each other
(+) = good agreement
(-) = poor agreement
84
What is a kappa interpretation and what are the values that are associated with it?
Used with a kappa statistic which is an agreement between evaluators
+1 = observers classify everyone the same way
0 = no relationship at all between the observers classifications above the agreement that would be expected by chance
-1 = observers classify everyone exactly opposite of each other
(+) = good agreement
(-) = poor agreement
85
What checklists are used for Interventional studies? (2)
**Consort Checklist -** CONsolidated Standards Of Reporting Trials
**PRISMA** Checklist - Preferred Reporting Items for Systemiatic reviews and Meta-Analyses
86
What type of study is the consort checklist used for? What type of trials is it used for?
**Consort** - randomized (clinical) trials = **interventional studies**
CONsolidated Standards Of Reporting Trials
Extension Documents - design extensions for **non-inferiority and equivalence trials**
**Cluster and Pragmatic Trials -** broadly defined as randomized, controlled trial whose purpose is to inform decisions about clinical practice. A philosophy as a continumm, not a dichotomy
87
What are the extension documents utilized for consort checklists for interventional studies?
Design extensions for non-inferiority and equivalence trials
Cluster Trials
Pragmatic Trials
88
What is a PRISMA checklist and what type of study is it used for?
Systematic reviews of multiple randomized trials
**P**referred **R**eporting **I**tems for **S**ystematic reviews and **M**eta-**A**nalyses
89
What type of checklist is utilized for observational studies; cohort, case-control, cross-sectional? (3)
**STROBE** - STrengthening the Reporting of OBservational studies in Epidemiology
**STROBE-ME -** Molecular Epidemiology Studies
**STREGA -** STrengthening the Reporting of Genetic Association studies
90
What type of studies are useful for the STROBE checklist?
STrengthening the Reporting of OBservational studies in Epidemiology
Useful for **Observational Studies; Case-control, Cohort and Cross-sectional**
91
What type of study is the STROBE-ME checklist useful for?
STregthening the Reporting of OBservational studies in Epidemiology-Molecular Epidemiology studies
Useful for Observational Studies; **Case-control, Cohort and Cross-sectional**
92
What type of study is the STREGA checklist useful for?
STrengthening the Reporting of Genetic Association studies
Useful for Observational Studies; **Case-control, Cohort and Cross-sectional**
93
What checklists are useful for Non-Randomized studies? (5)
**GRIPS** - Genetic Risk Prediction Studies
**QUADAS-2 -** QUality Assessment of studies of Diagnostic Accuracy in Systematic reviews
**REMARK -** REporting recommendations for tumor MARKer prognostic studies
**STARD -** STAndards for the Reporting of Diagnostic accuracy studies
**TREND -** Transparent Reporting of Evaluations with Non-randomized Designs (large populations)
94
What is a TREND checklist and what is it useful for?
**T**ransparent **R**eporting of **E**valuations with **N**on-randomized **D**esigns
Useful for **non-randomized studies**
95
What is the REMARK checklist and what type of study is it useful for?
**RE**porting recommendations for tumor **MARK**er prognostic studies
Useful for **non-randomized** studies
96
What is a GRIPS checklist and what type of study is it useful for?
**G**enetic **R**isk **P**rediction **S**tudies
Useful for **non-randomized** studies
97
What is a STARD checklist and what type of study is it useful for?
**STA**ndards for the **R**eporting of **D**iagnostic accuracy studies
Useful for **non-randomized** studies
98
What is a QUADAS-2 checklist and what type of study is it useful for?
**QU**ality **A**ssessment of studies of **D**iagnostic **A**ccuracy in **S**ystematic reviews
Useful for **non-randomized** studies
99
Where exactly can each checklist that is useful for all studies be located for future use?
**Equator Network** website
100
What is NCT and what is its purpose and where can it be found?
**Number of Clinical Trials**
A number assigned by **clinicaltrials.gov** once research protocol is submitted **prior** to study initiation
The purpose is to **reduce publication bias**
101
What type of data can be located in clinicaltrials.gov and how was it developed?
ICMJE-acceptable public registry that offers **up to date** data info for locating **interventional studies that are starting/in progress**
NIH developed site with the FDA through its national library of medicine (NLM)
