Using AI to Draft Study Preregistrations

End-of-lesson check

15 questions · take it digitally for instant feedback at tendril.neural-forge.io/learn/quiz/end-research-ai-preregistration-drafting-creators

1. What is a primary capability of AI when drafting a study preregistration?

   1. Guaranteeing that the study will be accepted by an open science registry
   2. Replacing the need for a methodologist to review the design
   3. Automatically collecting and analyzing research data
   4. Structuring hypotheses operationally and specifying analysis steps in advance

2. What does it mean that a preregistration is a 'timestamped commitment'?

   1. It can be modified at any time without consequences
   2. It automatically becomes invalid after the study is published
   3. It only matters for studies that receive funding
   4. It records the exact date and time when the research plan was formally recorded before data collection

3. Which of the following is explicitly listed as something AI CANNOT do in the preregistration process?

   1. Generate operational definitions for study variables
   2. Format the preregistration document according to registry guidelines
   3. Replace methodological expertise
   4. Suggest appropriate statistical tests for the research design

4. Why is it important to have a methodologist review a preregistration before submission?

   1. To verify that the planned methods can actually test the stated hypotheses
   2. To ensure the document is formatted correctly for the registry
   3. To translate the document into the required language
   4. To automatically submit the document to multiple registries at once

5. What type of information is typically included in an OSF-style preregistration?

   1. A detailed budget for the research project
   2. The hypotheses, design, measures, sample size, and planned analyses
   3. A list of previously published papers by the researcher
   4. The final study results and conclusions

6. A student wants to use AI to draft their psychology study's preregistration. What should they do before submitting it to a registry?

   1. Post it on social media for feedback from non-researchers
   2. Have a methodologist review it to ensure methodological soundness
   3. Submit it immediately since AI drafted it correctly
   4. Remove all technical language to make it simple

7. What is the main purpose of preregistering a study?

   1. To create a timestamped record of planned methods before data collection, reducing bias
   2. To guarantee funding approval from agencies
   3. To automatically publish the results in a journal
   4. To secure intellectual property rights over the research idea

8. A researcher discovers after collecting data that their original analysis plan won't work. What does the lesson recommend?

   1. Submit an amended preregistration explaining any changes and why they were necessary
   2. Continue with the original plan regardless of its validity
   3. Modify the preregistration to match what they actually did
   4. Delete the original preregistration and create a new one

9. Which statement best describes AI's role in the preregistration workflow?

   1. AI determines whether the research is scientifically valuable
   2. AI can fully replace human researchers in creating and validating preregistrations
   3. AI can help draft the document but cannot guarantee its acceptance by registries
   4. AI automatically submits preregistrations to all major registries

10. What distinguishes a hypothesis from an operational definition in research planning?

    1. Operational definitions are only needed for dependent variables
    2. Hypotheses are not necessary for preregistration
    3. They are essentially the same thing
    4. A hypothesis is a prediction, while an operational definition specifies how variables will be measured

11. Why might a registry reject a submitted preregistration?

    1. Because AI was used to draft it
    2. Because the study is too short
    3. Because the researcher has too many publications
    4. Because the methodology is unclear, incomplete, or inappropriate for testing the hypotheses

12. What does the lesson mean by 'specifying analysis steps in advance'?

    1. Waiting until after collecting data to decide on analyses
    2. Hiring a statistician after the study is complete
    3. Collecting all data immediately after stating the hypothesis
    4. Deciding which statistical tests will be used before seeing the data

13. What is 'p-hacking' and why is preregistration relevant to it?

    1. P-hacking occurs when participants hack into research databases
    2. Preregistration makes p-hacking easier
    3. P-hacking is manipulating analyses to find significant results; preregistration prevents this by locking in analysis plans beforehand
    4. P-hacking is a type of computer security issue; preregistration is unrelated

14. A student asks AI to generate a preregistration for their science fair project. What should they verify before submitting?

    1. That the AI will take credit for the project
    2. That the AI used the newest version of any registry template
    3. That the sample size is realistic and the measures are appropriate for their population
    4. That the preregistration will guarantee their project wins

15. When drafting a preregistration, what is the purpose of specifying a 'sampling plan'?

    1. To determine the order of questions in a survey
    2. To describe how participants will be recruited and how many are needed for the study
    3. To select which researchers will work on the project
    4. To decide which participants will be included in the final analysis