Brainscape's Knowledge GenomeTM


Top Flashcards (Certified)
All Flashcards

From Flashcardlet > Research > Flashcards

Research Flashcards

(96 cards)

Start Studying
0
Q

Alternate Hypothesis (Ha or H1)

A

Statement that population parameter has a value different from null hypothesis
Accepted when null hypothesis is rejected

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

Null Hypothesis (Ho)

A

The Statistical Hypothesis
Statement that the value of a population parameter (mean, proportion or correlation coefficient) is equal to a claimed value
Tested statistically by interferential statistics

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

Independent Variable

A

What caused or influenced dependent variable

What is controlled or manipulated

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

Dependent Variable

A

The response or outcome

Caused by effect of independent variable

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

P-Value

A

Probability a statistical result happened by chance

If smaller than alpha level, null hypothesis is rejected

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

Alpha Level

A

The significance level
Probability of rejecting the null hypothesis or chance of Type I error
Often a = 0.05 or 0.01

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

Type I Error (Alpha Error)

A

Wrongly decide to reject null hypothesis
Conclude there is a difference or relationship when there is not
If alpha = 0.01, there is 1% chance of this error
A false positive finding

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

Type II Error (Beta Error)

A

Wrongly decide not to reject null hypothesis
Conclude no difference or relationship when there is one
A false negative finding

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

Statistically Significant

A

Small probability the difference or relationship between groups/variables happened by chance

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

Statistical Power

A

Chance a statistical test will lead to rejecting a false null hypothesis (find a statistically significant result)

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

Effect Size (ES)

A

Measure of magnitude of difference or relationship between treatments/variables
Larger ES = Greater chance to be statistically significant

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

Effect Size Index

A

Represents ES using standardized value
Difference between 2 groups divided by standard deviation of 1 group
TREATMENTmean - CONTROLmean / TREATMENTsd or CONTROLsd
< 0.1 = trivial effect
0.1-0.3 = small effect
0.3-0.5 = moderate effect
> 0.5 = large effect

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

Minimally Clinically Important Difference (MCID)

A

Minimally clinically significant difference (MCSD)

Smallest difference considered worthwhile and warrant change in patient management

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

Minimal Detectable Change (MDC)

A

Minimal Detectable Difference (MDD)
Minimum detectable change in patient condition beyond threshold of measurement error
Smallest difference or change that is statistically significant
Standard error of measurement (SEM) is used to determine this amount

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

Parameter

A

Numerical measurement describing population characteristic
Greek letters
Mu for mean
Sigma for std dev

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

Statistic

A

Numerical measurement describing characteristic of a sample
English letters
x or M for sample mean
s for std dev

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

Forest Plot (Blobbogram)

A

Used in meta-analysis
Results of individual studies and cumulative summary of all studies
Square on Horizontal Line: Estimate of measure of effect (odds ratio, relative risk)
Line Width: Confidence interval
Square Area: Study’s weight in the meta-analysis
Diamond on Horizontal Line: Estimate of cumulative effect
Vertical Line: No effect or null value

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

External Validity

A 
Degree results are generalizable to populations or circumstances outside of study
Threats: 
  Treated specific type of subjects
  Place (setting) of study
  Time (history) of study
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Internal Validity

A 
Degree intervention caused outcome
Degree independent variable caused dependent variable
Threats:
  History
  Maturation
  Attrition
  Testing
  Instrumentation
  Regression towards the mean
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Hawthorne Effect

A

Unstated subject experiences change from being in study

Change of behavior due to being observed or studied

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

Placebo Effect

A

Inactive treatment causes improvement because patient has expectation it will help

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

Quasi-Experimental Design

A

Design without control group, randomization, or both

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

One-Group Pretest-Posttest Design

A

Measurements on one group before and after

Time is independent variable

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

One-Way Repeated Measures Design Over Time

A

Measurements on one group at multiple prescribed time intervals
Intervention may be done once or repeated

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Time Series Design
Multiple measurements before and after to observe patterns or trends
25
Sampling Error
Chance difference between sample statistic and true value of the population
26
Sampling
With Replacement: Each unit selected from population is replaced before drawing another Each unit has truly equal chance of being selected throughout sampling Rarely used on humans Without Replacement: Units selected are not returned to sample Cannot be selected twice for same sample Population size decreases as sample size increases
27
Simple Random Sampling
Random number generator Not most statistically efficient method May not yield representative sample - luck of the draw
28
Systematic Sampling
Every nth subject Interval based on population and desired sample size Simple
29
Stratified Random Sampling
Divided into strata Simple random sampling from each strata Representative of key subgroups
30
Cluster Sampling
Divided into clusters Random sampling of clusters Less costly, more efficient especially wide geographical area
31
Crossover Design
Subject receives each treatment (treatment and control) in random order separated by period of no treatment Subjects serve as own control
32
Factorial Design
Two or more independent variables | Different subjects assigned to the various combinations of independent variables
33
Matching/Pairing
Identify similar pairs of subjects prior to randomization to ensure balanced groups
34
Intention-To-Treat Analysis
All subjects randomly assigned to one of the treatments are analyzed together regardless of receiving or completing treatment Preserves original balance of groups
35
Non-Probability Sampling
Convenience: Readily available Purposive: Deliberately selected Quota: Convenience sampling of strata Snowball: Subjects identify other possible subjects
36
Kurtosis
"Peakedness" of a distribution Peaked or flat relative to a normal distribution High Kurtosis: Sharper peak with longer, flatter tails Low Kurtosis: Rounded peak with shorter, thinner tails
37
Normal Distribution
``` 68% = Values within 1 std dev 95% = Values within 2 std devs 99% = Values within 3 std devs ```
38
Skewness
``` Negatively Skewed: Mean and median left of mode Left tail elongated Positively Skewed: Mean and median right of mode Right tail elongated ```
39
Sensitivity
Percentage who test (+) in a group of people with the disease or condition
40
Specificity
Percentage who test (-) in a group of people without the disease or condition
41
False Negative Test
Indicates person does not have condition or disease when they actually do
42
False Positive Test
Indicates person does have condition or disease when they actually do not
43
Negative Predictive Value
Ability to correctly determine proportion of patients without condition/disease from all patients who test (-)
44
Positive Predictive Value
Ability to correctly determine proportion of patients with condition/disease from all patients who test (+)
45
Coefficient Of Variance (CV)
Ratio of the standard deviation of a distribution to the mean CV = (s/mean) x 100
46
Inferential Statistics
Use sample data to make inferences about a population Used to test hypotheses Parametric and nonparametric
47
Parametric Statistics
Assume samples from populations normally distributed and homogeneity of variance Interval and ratio data
48
ANOVA
Test equality of means between 2+ populations by analyzing sample variances
49
One-Way ANOVA
Compare the means of two or more populations Only one independent variable is examined One-Way: Sample data separated into different groups based on one characteristic or factor
50
Two-Way ANOVA
Compare the means of two or more populations Two or more independent variables are examined Two-Way: Sample data separated into different groups based on two characteristics or factors
51
Repeated Measures ANOVA
ANOVA where all subjects are measures under number of different experimental conditions Used for where practice or carryover effects are minimal Used for individuals matched according to important characteristic
52
Regression Analysis
Examine relationship between a dependent variable (Y) and 1+ independent predictor variables (X) Predicts how change in 1+ independent variables affects dependent variable
53
Confidence Interval (CI)
Range of values used to estimate a population parameter
54
Confidence Level
Probability the CI actually contains the unknown population parameter 95% CI = 95/100 times the sample will contain true population parameter
55
Point Estimate
Single value calculated from a sample that best approximates a population parameter Sample mean = point estimate for population mean
56
Intraclass Correlation Coefficient (ICC)
Assesses both degree of correspondence and agreement among scores Ranges from 0.0 - 1.0
57
Pearson Product Moment Correlation (r)
Measures magnitude and direction of linear relationship between 2 variables on the interval scale Ranges from -1.0 to +1.0 Sign = direction of relationship Number = Magnitude of relationship Value = Exactly -1.0 or +1.0 = Values fall in a straight line Value = 0 = No relationship
58
T-Test
Estimates population mean or compares 2 means when population normally distributed and variance is unknown
59
Dependent Or Paired T-Test
Compare means of 2 groups that are correlated | Paired sample t-test used when samples are matched pairs
60
Independent T-Test
Compare means of 2 independent groups Independent = Members of groups are different i.e. compare men and women because one could not be in both groups
61
One Sample T-Test
Compare a sample mean to a population expected or reference mean Used to decide if sample mean is different from the population
62
One-Tailed T-Test
Deviations from null hypothesis in only one direction are considered 0.05 Significance level: Apportions 0.05 to the right or left tail Only applies when test statistic is symmetrically distributed Implies intervention could only have one effect (beneficial or harmful)
63
Two-Tailed T-Test
Deviations forum null hypothesis in both directions are considered 0.05 Significance level: Apportions 0.025 to both tails Only applies when test statistic is symmetrically distributed
64
Z-Test
Estimating the population mean or comparing 2 means when population is normally distributed and variance is known
65
Nonparametric Statistics
Do not assume samples are from populations normally distributed with homogeneity of variance Nominal or ordinal data Interval or ratio data that are not normal
66
Chi-Square Test
Evaluates difference between observed and expected frequencies Examines association or independence between categorical variables Used with nominal data
67
Kruskal-Wallis Test
Determines if 3+ independent samples come from the same population No parametric version of one-way ANOVA
68
Mann-Whitney Test
Compares 2 independent samples with ordinal data Nonparametric alternative of independent t-test Equivalent to the Wilcoxon signed rank test
69
Spearman Rank Correlation Coefficient
Correlation test for association between two variables Ranked (ordinal) data - can be converted to ranks if not originally Nonparametric equivalent to Pearson product moment correlation Ranges from -1.0 to +1.0
70
Wilcoxon Signed Rank Test
Compares 2 dependent samples with ordinal data | Nonparametric alternative of the dependent or paired t-test
71
Incidence
Number of NEW cases in population at risk during a specified TIME INTERVAL NEW CASES per 100,000 people at risk
72
Prevalence
Number of EXISTING CASES at A POINT IN TIME Includes new and pre-existing cases EXISTING CASES per 100,000 people at risk
73
Relative Risk
The Risk Ratio Measure of risk of certain event happening in one group compared to another Incidence among individuals exposed to risk factor : incidence among individuals not exposed to risk factor RR Risk 1.0 = Event equally probable in both groups RR Risk > 1.0 = Exposure increases risk RR Risk < 1.0 = Exposure decreases risk
74
Odds Ratio (OR)
The Relative Odds Measure of the odds of event happening in one group compared to another Most often in case control (backward-looking) studies to see if risk factor increases risk of developing disease OR 1.0 = Exposure probably does not increase risk OR > 1.0 = Exposure may increase risk OR < 1.0 = Exposure may decrease risk
75
Continuous Data
Continuous scale, range of values without gaps or interruptions ROM, distance, weight, time
76
Discrete Data
Whole units | HR, patients diagnosed with CA, number of PT visits
77
Dichotomous Data
Limited to 2 values | Male or female, yes or no
78
Qualitative Data
Categorical data | Eye color, blood type, hand dominance
79
Quantitative Data
Numbers that represent counts or measurements | Numeral assigned an object, event, or person
80
Nominal Scale
Category | Blood type, BS, arthritis type
81
Ordinal Scale
Ranking | MMT, levels of assistance, pain, joint laxity grades
82
Interval Scale
Intervals between adjacent values are equal but no true zero point Temperature
83
Ratio Scale
Intervals between adjacent values are equal with true zero point ROM, distance walked, time to complete activity, NCV
84
Reliability
Reproducibility or repeatability of measurements
85
Alternate Forms Reliability
Parallel Forms Reliability Consistency or agreement obtained with different forms of a test SAT, GRE, NPTE can be administered as long as they are equivalent measures
86
Internal Consistency
Extent to which items or elements contributing to measurement reflect one basic phenomenon or dimension Items on functional assessment scale relate to patient's function
87
Validity
Degree to which useful or meaningful interpretation can be inferred from a measurement
88
Face Validity
Degree to which measurement appears to test what it is supposed to test
89
Content Validity
Degree to which measure,eat reflects meaningful elements of construct and items in test reflect content domain of interest McGill pain questionnaire > analog pain scale - Not just intensity, also location, quality, and duration of pain
90
Construct Validity
Degree to which theoretical construct in measured | MMT = indicator of Innervation status of muscle if relationship between score and EMG testing
91
Criterion-Related Validity
Established by comparing to gold standard or data obtained by different forms of testing
92
Concurrent Validity
Comparing to gold standard
93
Predictive Validity
Measurement is predictive of future behavior or event
94
Prescriptive Validity
Measurement suggests treatment patient should receive
95
Hierarchy/Levels Of Evidence
1. Systematic Reviews, Meta-Analysis 2. RCTs 3. Cohort Studies 4. Case Control Studies 5. Cross-Sectional Studies 6. Case Series 7. Case Report 8. Ideas, Opinions

From Flashcardlet (74 decks)

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