LAISHA
Open Laisha

ASA Accredited Professional Statistician (PStat) and Graduate Statistician (GStat)

USA · American Statistical Association

30practice questions written to this syllabus
3subjects

What ASA Accredited Professional Statistician and Graduate Statistician is

ASA Accredited Professional Statistician (PStat) and Graduate Statistician (GStat) is a professional certification run by American Statistical Association, for candidates in USA.

Format

Portfolio-based accreditation, not a written exam: advanced degree, documented experience, references and commitment to ethics and continuing education.

Sections

How it has changed

Read on the ASA accreditation page. Used by statisticians in consulting, government and industry, including agricultural research.

How to prepare for ASA Accredited Professional Statistician and Graduate Statistician

Where our 30 questions for it fall

Our reading of the syllabus, from the questions we have written to this exam; the exam body's own weighting, where published, is under Format above.

  1. Heaviest (50% of the questions): Statistical Consulting, Research Methodology & Scientific Communication (15).

    Start here: read these subjects' manuals first and work their questions until the reasoning is automatic.

  2. Middle band: Data Analysis with Statistical Software & Reproducible Research (8).

    One pass per subject, then mixed sets across them.

  3. Lighter: Statistical Computing: Simulation, Resampling & Algorithms (7).

    Definitions, classifications and standard figures; quick to revise, easy marks to lose.

  4. Hardest share: Data Analysis with Statistical Software & Reproducible Research (1 hard of 8), Statistical Consulting, Research Methodology & Scientific Communication (1 hard of 15).

    Leave untimed time for these and read every explanation, including for the options you rejected.

Official source: amstat.org

8 practice questions, worked

Original questions written to this exam's published syllabus and pattern. Not past papers.

1. Checking a key analysis in a second program mainly helps to:

  1. Double the p-value
  2. Hide discrepancies
  3. Reveal specification errors
  4. Avoid reporting versions
Answer and working

C. Reveal specification errors

Agreement suggests the model was specified correctly.

Data Analysis with Statistical Software & Reproducible Research · Genstat and specialist software for agricultural research

2. An observed coverage of 0.94 from 1,000 replicates for a nominal 95% interval is:

  1. Not clear evidence of under-coverage
  2. Proof of serious under-coverage
  3. Proof of over-coverage
  4. Impossible
Answer and working

A. Not clear evidence of under-coverage

The Monte Carlo SE is about 0.007, so 0.94 is within 1.5 SE of 0.95.

Statistical Computing: Simulation, Resampling & Algorithms · Simulation studies: design and reporting

3. ARRIVE guidelines concern the reporting of:

  1. Household surveys
  2. Software code
  3. Weather data
  4. Research using animals
Answer and working

D. Research using animals

They list essential items for animal studies.

Statistical Consulting, Research Methodology & Scientific Communication · Reporting standards for experiments and surveys

4. When data cannot answer a manager's question, the statistician should:

  1. Say so and explain what is needed
  2. Give an answer anyway, to keep the manager satisfied
  3. Hide the limitation
  4. Change the question without telling them
Answer and working

A. Say so and explain what is needed

Overconfidence damages credibility.

Statistical Consulting, Research Methodology & Scientific Communication · Presenting statistical results to non-statisticians

5. A merge conflict occurs when:

  1. A new file is added to the working directory
  2. Branches edit the same lines differently
  3. A commit has a message
  4. The repository is cloned
Answer and working

B. Branches edit the same lines differently

The user edits the file to resolve it and commits.

Data Analysis with Statistical Software & Reproducible Research · Version control with Git

6. Before parallelising slow statistical code, it is usually best to:

  1. Buy more cores immediately
  2. Profile and vectorise it
  3. Remove all random numbers
  4. Rewrite it in assembly language
Answer and working

B. Profile and vectorise it

Algorithmic improvements often beat extra hardware.

Statistical Computing: Simulation, Resampling & Algorithms · Parallel and high-performance computing for statistics

7. 'Disease fell from 20% to 10%' is clearer than '50% reduction' mainly because it:

  1. It uses a larger number, which audiences prefer
  2. Hides uncertainty
  3. Avoids percentages
  4. It gives absolute levels
Answer and working

D. It gives absolute levels

Relative changes alone can mislead.

Statistical Consulting, Research Methodology & Scientific Communication · Presenting statistical results to non-statisticians

8. Restating the problem in your own words and asking the scientist to confirm mainly:

  1. Saves computing time
  2. Prevents misunderstandings
  3. Replaces the analysis plan
  4. Shows the statistician is senior
Answer and working

B. Prevents misunderstandings

Mismatched understanding otherwise surfaces only after the analysis.

Statistical Consulting, Research Methodology & Scientific Communication · The consulting interview: from scientific question to statistical question

Last updated · How Laisha writes and checks its pages · Report a mistake

Ask Laisha about this

Opens Laisha's chat with your question ready to send. Sign in or start free.

Unlock all 30 questions — ASA Accredited Professional Statistician (PStat) and Graduate Statistician (GStat)

US$10 a month · one price worldwide · cancel any time

  • Every practice question for this course, with explanations
  • Timed papers for the exams this course leads to
  • Ask Laisha to explain any topic — up to 1,000 questions a month
Free or Smarter
FreeSmarter
Grow guides, pest pages, articlesRead allRead all
Ask Laisha5 a dayUp to 1,000 a month
Exam practiceThe sample on each pageEvery question, timed papers
Crop calendar and pest alerts—For your own fields
Calculators and tools—All
Uploads and reports—100 a month, real files
10 days free for every new account, no card.