Research 2 Flashcards by Brooke Routhier

type I error

false positive

How well did you know this?

Not at all

Perfectly

type II error

false negative

How well did you know this?

Not at all

Perfectly

Sensitivity

Ability of a diagnostic test to identify true disease without missing anyone by leaving the disease undiagnosed

How well did you know this?

Not at all

Perfectly

Highly sensitive test:

̶Few false negatives

̶Helps in ruling conditions “out” (SnOut)

How well did you know this?

Not at all

Perfectly

sensitive negative test =

rule out the disease

How well did you know this?

Not at all

Perfectly

sensitive positive test =

presence of the disease

How well did you know this?

Not at all

Perfectly

Specificity

Ability of a diagnostic test to be correctly negative in the absence of disease without mislabeling anyone

How well did you know this?

Not at all

Perfectly

Highly specific test:

Few false positives
̶
Helps in ruling conditions “in” (SpIn)

How well did you know this?

Not at all

Perfectly

specific positive test =

rule in the disease

How well did you know this?

Not at all

Perfectly

specific negative test =

absence of the disease

How well did you know this?

Not at all

Perfectly

A physical therapist is using a new special screening test to help
determine if a tennis player has lateral epicondylitis among other possibilities. The test has a specificity of 0.91. What can be concluded if the results of the test are positive?
A. Try another special test for lateral epicondylitis
B. Consider the test results inconclusive
C. Rule in lateral epicondylitis
D. Rule out lateral epicondylitis

C. Rule in lateral epicondylitis

How well did you know this?

Not at all

Perfectly

test positive and has disease =

true positive

How well did you know this?

Not at all

Perfectly

test positive and does not have disease =

false positive

How well did you know this?

Not at all

Perfectly

test negative and has disease =

false negative

How well did you know this?

Not at all

Perfectly

test negative and does not have disease =

true negative

How well did you know this?

Not at all

Perfectly

Sensitivity is:

TP/(TP+FN)

How well did you know this?

Not at all

Perfectly

Specificity is:

TN/(TN+FP)

How well did you know this?

Not at all

Perfectly

A researcher is collecting data on 100 patients with venous insufficiency. Out of the 100 patients with venous insufficiency, 70
patients had a positive doppler test and 30 patients had a negative
doppler test. What is the sensitivity of the test?
A. 20%
B. 80%
C. 40%
D. 70%

D. 70%

How well did you know this?

Not at all

Perfectly

Dependent Variable:

Is the outcome or variable of interest in an experiment or study

patient outcome

How well did you know this?

Not at all

Perfectly

Independent Variable:

Variable that is manipulated or changed
by the researcher to observe its effect on the dependent variable

PT treatment

How well did you know this?

Not at all

Perfectly

Null Hypothesis:

No significant differences in group means or no
relation between the two variables

How well did you know this?

Not at all

Perfectly

Alternative Hypothesis:

Significant differences in group means

How well did you know this?

Not at all

Perfectly

Alpha value:

Predefined threshold that determines if your result is
statistically significant

Level of statistical significance, usually set before the study

If set at 0.05 à allowing 5% chance of error

How well did you know this?

Not at all

Perfectly

p-value:

Result of your statistical test

p value of 0.05 indicates 5% probability that the difference between the
groups is due to chance

How well did you know this?

Not at all

Perfectly

For a study to be statistically significant:

p value must be lower or equal to alpha level

A study examining the effect of dry needling on low back pain uses an alpha level of 0.05 and has a p-value of 0.02. What should the researcher conclude? A. Fail to reject the null hypothesis; the result is not significant. B. Reject the null hypothesis; the result is statistically significant. C. The result is clinically significant but not statistically significant. D. The result is inconclusive

B. Reject the null hypothesis; the result is statistically significant.

Parametric research

Equal distribution Randomization of sample Bell shaped curve: Normal distribution Quantitative data (Usually interval, ratio) More powerful

Non-Parametric research

Unequal distribution No randomization of sample Skewed curve: Non normal distribution Qualitative data (Usually nominal, ordinal) Less powerful than parametric

Parametric Test

t-test Analysis of Variance (ANOVA)

t-test

independent samples paired 1 tailed 2 tailed

Independent Samples t test:

Compares the difference between 2 independent groups (between subjects)

Paired t Test:

Compares difference between 2 matched groups investigate whether there's a difference within a group between 2 points in time (within-subjects)

1 tailed:

Directional hypothesis, 1 end of distribution, either +ve or –ve ̶ Ex. Pain status of the patient: Pre-PT > Post-PT

2 tailed:

Non-directional, 2 ends of distribution, both +ve and –ve ̶Ex. Effect of a novel intervention on pain could be Pre-PT > Post-PT OR Pre-PT < Post-PT

Analysis of Variance (ANOVA)

α- One-way ANOVA β- Two-way ANOVA γ-Repeated measures ANOVA

α- One-way ANOVA:

At Least 3 or more independent groups compared on 1 intervention

β- Two-way ANOVA:

At Least 3 or more independent groups compared on 2 intervention Ex. effects of both gender and exercise type (strength vs aerobic) on cardiovascular fitness

γ-Repeated measures ANOVA:

Individuals measured over time. Ex. effect of three different diets (Diet A, Diet B, and Diet C) on weight loss measured at baseline, 2 weeks and 4 weeks

A physical therapist is looking at squatting activity in 30 individuals who are overweight and 30 individuals who are at normal weight. Both groups performed 10 bilateral squats at 100-degrees of knee flexion. The physical therapist hypothesizes that knee moments will be higher in overweight. Which is the MOST APPROPRIATE test? A. One way ANOVA B. Two-way ANOVA C. One tailed t-test D. Two tailed t-test

C. One tailed t-test

ANCOVA:

Extension of ANOVA Examine significant differences among the means of two or more groups while statistically controlling for the influence of covariates Example: PT is testing tall and short subjects for gait using assistive devices. Here height is the covariate that must be controlled during statistical analysis since tall subjects will have greater step length compared to short subjected

Covariate:

Considered as an independent variable in an analysis but is not of primary interest. It is a potential source of variation or confounding factor that potentially influences the dependent variable

Non-Parametric Tests

Chi square test Mann Whitney U test Kruskal Wallis test

Chi square test:

Use of nominal/categorical data to find difference between groups * E.g., 20 males vs 30 females

Mann Whitney U test:

Use of continuous or ordinal data to test the null hypothesis with two independent samples from the same population * Similar to Independent samples t-test

Kruskal Wallis test:

Three or more groups compared * Similar to One-way ANOVA

A study is looking at effects of different exercise options for a patient with knee osteoarthritis. The study compared pool exercise, treadmill walking, and ankle weights on knee pain in three equal groups of 30 adults each. What is an appropriate statistical test to assess the effectiveness of treatment in three groups? A. K Wallis test B. Two paired t-tests C. ANOVA D. Chi square test

C. ANOVA

Correlation

Describes the strength and direction of a relationship between two variables.

Positive Correlation:

Both variables show same direction of change Ex. Number of hours studying and NPTE scores

Negative Correlation:

One variable increases while the other decreases. Ex. Duration of exercise and body weight

Correlation Strength:

Indicates how strongly two variables are related [positive from 0 – 1.00 negative from -1.00 – 0 ] * High > .76 to 1.00 * Moderate .51 to .75 * Fair .26 to .50 * Low 0.00 to .25

In a research study on patients who are obese with total hip arthroplasty, a correlation coefficient (r) of +0.80 was found for the relationship between weight and BMI. What is the MOST APPROPRIATE interpretation of this finding? A. Weight and BMI have a high positive correlation B. 80 percent of the variability in BMI can be accounted for by weight C. 80 percent of the variability in weight can be accounted for by BMI D. There are no significant differences between the weight and BMI levels

A. Weight and BMI have a high positive correlation

Regression

examines how changes in the independent variables (predictors) are associated with changes in the dependent variable (outcome) purpose is to generate an equation that relates X to Y, such that if given values of X, Y can be predicted Predicting NPTE scores based on PEAT scores Expressed in the form of an equation

Expressed in the form of an equation, R square ___ indicates strong association between variables

>0.5

A physical therapist collected data on patients with total hip arthroplasty and plotted the relationship of trunk flexion with waist circumference. The regression plot showed a R square value of 0.64 (See picture). What is the best way to describe the slope? A. Weak positive slope B. Weak negative slope C. Strong positive slope D. Strong negative slope

D. Strong negative slope

Parametric Test Normal Distribution Equal Samples

Independent T-test Paired T-test ANOVA Pearson product correlation

Non-Parametric Test Skewed Distribution Unequal Samples

Mann-Whitney U-test Wilcoxon Signed Rank test Kruskal Wallis test Spearman rho correlation

Purpose of the Test: Compares mean between 2 independent groups

Independent T-test Mann-Whitney U-test

Purpose of the Test: Compares mean between 2 dependent groups

Paired T-test Wilcoxon Signed Rank test

Purpose of the Test: Compares mean between 3 or more independent groups

ANOVA Kruskal Wallis test

Purpose of the Test: Quantifies association between 2 variables

Pearson product correlation Spearman rho correlation

Research 2 Flashcards

(60 cards)