OCPS Research Methods & Evidence-Based Practice 3 — Questions and Answers
Question 1: A payroll director notices that error rates spiked after hiring new staff but also after a software update. The inability to determine which factor caused the spike is known as:
- Sampling error
- Confounding (Correct answer)
- Random variation
- Type I error
Correct answer: Confounding
Confounding occurs when two variables change simultaneously, making it impossible to isolate which one caused the observed outcome.
Question 2: Which statistical measure best describes the spread or variability of payroll processing times across a department?
- Mean
- Mode
- Standard deviation (Correct answer)
- Median
Correct answer: Standard deviation
Standard deviation quantifies how much individual processing times deviate from the average, indicating variability.
Question 3: A payroll researcher finds a p-value of 0.03 when testing whether two payroll systems differ in accuracy. This result means:
- There is a 3% chance the systems are equally accurate
- There is a 97% chance the null hypothesis is true
- There is a 3% probability of observing the result if no true difference exists (Correct answer)
- The difference is practically significant for business purposes
Correct answer: There is a 3% probability of observing the result if no true difference exists
A p-value represents the probability of obtaining the observed result (or more extreme) assuming the null hypothesis is true.
Question 4: In evidence-based payroll practice, 'external validity' concerns whether:
- The study was approved by an external auditor
- Findings from one payroll environment can be applied to other organizations (Correct answer)
- The researchers were from outside the company
- The data was collected by a third party
Correct answer: Findings from one payroll environment can be applied to other organizations
External validity (generalizability) refers to how well research findings apply beyond the original study context.
Question 5: A payroll team uses a Likert-scale survey to assess employee satisfaction with pay stub clarity. This instrument generates which type of data?
- Nominal
- Ratio
- Ordinal (Correct answer)
- Continuous interval
Correct answer: Ordinal
Likert scales produce ordinal data because responses have a meaningful order but unequal intervals between points.
Question 6: Which approach to literature review is considered most rigorous and least prone to researcher bias?
- Narrative review based on personal expertise
- Systematic review with predefined inclusion/exclusion criteria (Correct answer)
- Scoping review limited to recent publications
- Integrative review combining all available sources
Correct answer: Systematic review with predefined inclusion/exclusion criteria
Systematic reviews follow a predetermined protocol with explicit criteria, reducing bias and improving reproducibility.
Question 7: When a payroll compliance study reports a confidence interval of [2.1%, 4.7%] for error rate reduction, this means:
- The true effect is exactly 3.4%
- We are 95% confident the true reduction falls between 2.1% and 4.7% (Correct answer)
- There is a 95% chance the study is correct
- The minimum expected improvement is 4.7%
Correct answer: We are 95% confident the true reduction falls between 2.1% and 4.7%
A 95% confidence interval means that if the study were repeated many times, 95% of calculated intervals would contain the true population value.
OCPS Research Methods & Evidence-Based Practice 3
A payroll director notices that error rates spiked after hiring new staff but also after a software update.
The inability to determine which factor caused the spike is known as: