BCPP Statistics PEDS Questions/Answer 2025/2026
Graded A+
Nominal variable - Discrete variable classified into groups in an unordered manner &
w/no indication of relative severity (male or female, dead or alive, disease presence or
not, race)
Ordinal variable - Discrete variable ranked in a specific order but with no consistent
level of magnitude of difference between ranks (NYHA functional classes)
Discrete variables - Can take only a limited # of variables within a given range (Nominal
& Ordinal)
Continuous variables - a second type of random variable (other than discrete), can take
on any value within a given range
Two types of continuous variables & their descriptions - 1. interval--data ranked in a
specific order with a consistent change in magnitude between units, zero point is
arbitrary (degrees Fahrenheit)
2. Ratio--HAS an absolute zero (heart rate, blood pressure, time, distance, otherwise
like interval)
Median - midpoint of values when placed in order from highest to lowest
Mode - most common value in a description
Two parameters that define a normally distributed population (Gaussian) - 1. mean
2. SD
95% CI calculation - mean +/- (2*SEM)
CI's - Help determine importance of findings, give an idea of magnitude of difference
between groups & the statistical significance
~a range data together w/a point estimate of the difference
What do p-values tell us? - tell if statistically significant difference between groups but
they don't tell us anything about the magnitude of the difference
CI that includes 0 - means no difference between two variables, interpreted as NOT
statistically significant (a p-value of 0.05 or greater)
(No need to show both 95% CI & the p-value.)
Assumptions of parametric tests - 1. data have an underlying normal distribution
(estimate by mean~median)
2. continuous data (either interval or ratio)
3. data have variances that are homogeneous between the groups investigated
, When to use Nonparametric test - 1. data aren't normally distributed
2. data don't meet other criteria for parametric tests (discrete data)
Parametric test for comparing sample with population mean - Student t-test one-sample
Nominal data test comparing expected & observed proportions between 2 or more
groups - Chi-square (x2)
Nominal test when data is less than 5 predicted observations - Fisher exact test
Nominal data paired samples test - McNemar
Nominal data test controlling for the influence of confounders - Mantel-Haenszel
Non parametric test for two independent samples - Wilcoxon rank sum, Mann-Whitney
U, Wilcoxon Mann-whitney
Non parametric test for 3 or more independent samples, - Kruskal-Wallis one-way
ANOVA by ranks
Nonparametric test for two matched or paired samples - Sign test & Wilcoxon signed
rank test
Non-parametric test for three or more matched or paired groups - Friedman ANOVA by
ranks
Type II error (beta error) - false negative
p-value - calculation that a Type I error has occurred
Pearson correlation - strength of relationship between 2 variables (r ranges -1 - 1)
Power - ability to detect differences between groups if one actually exists (1-Beta)
Sensitivity - proportion of patients w/disease who have a positive test (percent of true
positives)
Specificity - proportion of patients without disease who have a negative test (percent of
true negatives)
calculation of sensitivity - True positives divided by False negatives + True positives
(A/(A+C))
(TP/(FN + TP))
Graded A+
Nominal variable - Discrete variable classified into groups in an unordered manner &
w/no indication of relative severity (male or female, dead or alive, disease presence or
not, race)
Ordinal variable - Discrete variable ranked in a specific order but with no consistent
level of magnitude of difference between ranks (NYHA functional classes)
Discrete variables - Can take only a limited # of variables within a given range (Nominal
& Ordinal)
Continuous variables - a second type of random variable (other than discrete), can take
on any value within a given range
Two types of continuous variables & their descriptions - 1. interval--data ranked in a
specific order with a consistent change in magnitude between units, zero point is
arbitrary (degrees Fahrenheit)
2. Ratio--HAS an absolute zero (heart rate, blood pressure, time, distance, otherwise
like interval)
Median - midpoint of values when placed in order from highest to lowest
Mode - most common value in a description
Two parameters that define a normally distributed population (Gaussian) - 1. mean
2. SD
95% CI calculation - mean +/- (2*SEM)
CI's - Help determine importance of findings, give an idea of magnitude of difference
between groups & the statistical significance
~a range data together w/a point estimate of the difference
What do p-values tell us? - tell if statistically significant difference between groups but
they don't tell us anything about the magnitude of the difference
CI that includes 0 - means no difference between two variables, interpreted as NOT
statistically significant (a p-value of 0.05 or greater)
(No need to show both 95% CI & the p-value.)
Assumptions of parametric tests - 1. data have an underlying normal distribution
(estimate by mean~median)
2. continuous data (either interval or ratio)
3. data have variances that are homogeneous between the groups investigated
, When to use Nonparametric test - 1. data aren't normally distributed
2. data don't meet other criteria for parametric tests (discrete data)
Parametric test for comparing sample with population mean - Student t-test one-sample
Nominal data test comparing expected & observed proportions between 2 or more
groups - Chi-square (x2)
Nominal test when data is less than 5 predicted observations - Fisher exact test
Nominal data paired samples test - McNemar
Nominal data test controlling for the influence of confounders - Mantel-Haenszel
Non parametric test for two independent samples - Wilcoxon rank sum, Mann-Whitney
U, Wilcoxon Mann-whitney
Non parametric test for 3 or more independent samples, - Kruskal-Wallis one-way
ANOVA by ranks
Nonparametric test for two matched or paired samples - Sign test & Wilcoxon signed
rank test
Non-parametric test for three or more matched or paired groups - Friedman ANOVA by
ranks
Type II error (beta error) - false negative
p-value - calculation that a Type I error has occurred
Pearson correlation - strength of relationship between 2 variables (r ranges -1 - 1)
Power - ability to detect differences between groups if one actually exists (1-Beta)
Sensitivity - proportion of patients w/disease who have a positive test (percent of true
positives)
Specificity - proportion of patients without disease who have a negative test (percent of
true negatives)
calculation of sensitivity - True positives divided by False negatives + True positives
(A/(A+C))
(TP/(FN + TP))
