Preparation to Statistics and Research Methodology

This set of vocabulary flashcards covers essential statistical concepts, research study designs, reporting guidelines (like STROBE and CONSORT), and diagnostic accuracy measures including sensitivity, specificity, and likelihood ratios.

STROBE statement

A reporting guideline used for observational studies, including cohort, case-control, and cross-sectional designs.

CONSORT

The standard reporting guideline used for randomised trials.

PRISMA

Reporting guidelines designed for systematic reviews.

What statistical test is most appropriate for examining the association between two categorical variables in a cross-sectional survey?

Chi-square test of association

Shapiro-Wilk test

A statistical test used in conjunction with histograms and Q-Q plots to check for the normality of continuous data.

Independent-samples t-test

A parametric test used to assess significant differences between two groups for continuous, normally distributed data.

Chi-square test of association

A statistical test used to assess significant differences or relationships between groups for categorical data.

Odds Ratio (OR)

A measure of association commonly used in case-control and cross-sectional studies; calculated as the odds of outcome in the intervention group divided by the odds in the control group.

Risk Ratio (RR)

A measure of association used in experimental and cohort studies; calculated as the risk of outcome in the intervention group divided by the risk in the control group.

Risk Difference

The absolute difference in risk between the intervention and control groups; calculated as $Risk_{control} - Risk_{intervention}$.

Risk (Definition)

The probability of having a specific event among a total population.

Sensitivity

The ability of a test to correctly identify people with the condition; calculated as $\frac{TP}{TP + FN}$.

Specificity

The ability of a test to correctly identify people without the condition; calculated as $\frac{TN}{TN + FP}$.

SnNout

A clinical mnemonic standing for: a highly Sensitive test, when Negative, rules the condition OUT.

SpPIn

A clinical mnemonic standing for: a highly Specific test, when Positive, rules the condition IN.

Positive Likelihood Ratio (LR+)

Indicates how much more likely a positive test result is to occur in a person with the condition than without; calculated as $\frac{sensitivity}{1 - specificity}$.

Negative Likelihood Ratio (LR-)

Indicates the likelihood of a negative result in someone with the condition; lower values indicate better performance in ruling out the condition, calculated as $\frac{1 - sensitivity}{specificity}$.

Fagan's nomogram

A chart used to determine the posterior (post-test) probability of a disease based on the prior (pre-test) probability and the likelihood ratio.

ANOVA

Analysis of variance; a statistical test used to compare the means of more than two groups.

Effect Size

A measure relating to the magnitude of the differences found in a study (e.g., Cohen's d, Odds Ratio, Pearson's r).

Meta-analysis

A statistical procedure in which the effect sizes from multiple individual studies are combined.

P-value

A value used to determine if findings are due to chance by evaluating the null hypothesis ($H_0$).

Skewness

A measure of the asymmetry of a probability distribution; categorized as 'positive' if the mean is greater than the median, and 'negative' if the mean is less than the median.

True/false Normality assumptions must be checked before using any parametric tests

True

The JBI tool for qualitative research emphasises examining BLANK and BLANK as part of the appraisal

Reflexivity and positionality

In a box plot, the BLANK represents the range of the middle 50% of the data

interquartile range

What is the function of the correlation coefficient?

measures the strength of relationship between two variables

Definition of continuous data

numerical data that can take any value within a range

Definition of discrete data

numerical data that can take only specific values, often counted in whole numbers.

Sensitivity

proportion of true positives identified by the test

Specificity

The proportion of true negatives identified by the test

Likelihood ratio +

The increase in odds of the condition being present when the test is positive

Likelihood ratio -

The decrease in odds of the condition being present when the test is negative

What is the significance of a specificity of 45% in this context?

indicates a relatively low ability to correctly identify patients without the condition

What is study ‘blinding’

keeping study participants unaware of group allocation

What is ‘allocation concealment’

Preventing study personnel from knowing the future assignment

What type of data is VO2 max measured in when assessing the affects of exercise interventions?

Continuous data

What is a ratio in continuous variables?

has a true, absolute zero point - CAN say one value is twice as much as another

What is an interval variable in continuous variables?

Has an arbitrary zero point that does NOT mean nothing or absence - cannot say Twice as much

Example of nominal data

eye colour

Example of ordinal data

level of satisfaction

Example of interval data

Temperature

Example of ratio data

Height

What does it mean if a test has a high LR+ and a low LR-

greater discriminating ability