Module 2 Practice Quiz

Replicability in Bioinformatics (Beginner check-in)

Answer all questions, then select Grade Quiz to view feedback.

What this quiz covers

What “replicability” means (and why the term can be confusing)
Why independent replication attempts matter for trust
Why replication can fail even when data and code are shared
How negative and null results fit into scientific progress

Each question has one best answer. Your results appear at the bottom of the page.

1) In this module, “replicability” is most closely described as:

A. Getting the same result with the same data and the same code B. An independent group obtaining the same result using artifacts they develop independently C. Making figures look identical across papers D. Publishing only statistically significant findings

2) Why does the module warn that “replicability” can be confusing?

A. Because different disciplines use different definitions for reproducibility/replicability B. Because computers always produce random outputs C. Because journals do not allow methods sections D. Because code can never be shared

3) The ACM definition emphasized in the slides suggests replicability involves:

A. The same team rerunning the same script on the same computer B. A different team, different setup, and independently developed artifacts C. Peer review only, with no re-running of analyses D. Publishing results without code to prevent misuse

4) Replication attempts are connected to “trust” because they help us:

A. Avoid sharing methods B. Decide whether findings hold up when independently tested C. Guarantee every result is true D. Replace the need for statistics

5) Even when data and code are provided, replication/reproduction can fail due to:

A. Software version differences, missing metadata, incomplete documentation, or human error B. Too many citations in the paper C. Using a PDF for the manuscript D. The presence of a methods section

6) Why can bioinformatics pipelines create many “opportunities for modification”?

A. Pipelines usually have multiple tools and parameters where choices can differ B. Because sequencing instruments always change file extensions C. Because code is never reused D. Because figures cannot be regenerated

7) A “negative result” most commonly means:

A. The study was unethical B. Results are not statistically significant (or the expected effect was not found) C. The data are always wrong D. The authors did not try hard enough

8) “Publication bias” in this context refers to:

A. Journals preferring to publish only positive or “exciting” findings B. Journals preferring to publish only long papers C. Journals preferring to publish only papers with fewer than 10 references D. Journals preferring papers with no figures

9) “Gray literature” generally means:

A. Evidence not published in commercial publications (often not formally peer reviewed) B. Articles printed in black-and-white only C. Any paper older than 10 years D. Any dataset without a DOI

10) The slides emphasize: “Failure to replicate is not failure to do good science” because it can:

A. Reveal meaningful differences, incorrect assumptions, or errors—and generate new questions B. Prove that science is pointless C. Eliminate the need for documentation D. Guarantee the original authors committed misconduct

What to do if you missed questions

Review the definition(s) of replicability and how they vary across disciplines.
Revisit examples of why replication can fail even with shared data/code.
Read the section on negative/null results and publication bias.