Designing assessment to prevent academic dishonesty

1
Designing assessment to prevent academic
dishonesty

• George Dunbar

2
Who cheats?

• Less mature people
• Men
• People with lower grades

3
But

• Its almost all self-report data
• Differences can be small, inconsistent, may be
  narrowing
• Multivariate studies wash some effects out

4
Prevention

• Teach students academic skills
• Tell students the ground rules
• Discuss cheating and plagiarism
• Set a good example
• Manage the timing and quantity of assessment
• Design courses and assessment to reduce
  opportunities to gain from cheating
• Monitor data collection
• Raise detection levels
• Punish in a consistent and transparent way

5
Explain

• Instruct students in academic skills
• Citation
• Quotation
• Paraphrase
• Note-taking
• Common ground
• Tell students the ground rules
• Discuss cheating and plagiarism

6
Set a good example

• Set a fair example
• Cyril Burt
• Turkish case
• Neil Winn
• Textbooks and Eskimos
• Eric Poehlman
• Take it seriously, take a balanced approach

7
Set standards

• Honor codes
• Place responsibilities on students
• Challenge academic dishonesty

8
51

• of academic staff admitted to suspecting
  cheating and ignoring it.
• Barrett Cox (in press)
• Students notice

9
Avoid creating

• Dont let a situation arise where students might
  feel the outcome would be unfair if they didnt
  cheat
• An unfairly hard exam, outcome dependent on
  luck
• Unfairly harsh or arbitrary marking

10
Manage assessment

• Manage the timing and quantity of assessment
• Design courses and assessment to reduce
  opportunities to gain from cheating
• Use oral presentations or oral exams
• Monitor data collection

11
More tips

• Timing matters
• Allow time
• Help students anticipate load
• Structure assessment
• several small
• outline first

12
Tips

• Ask for drafts or annotated bibliographies
• Cut out online sources
• choose titles carefully
• require reference to recent papers/events
• require localisation of work
• require unusual genres or lengths

13
Abolish coursework

• Use exams only the Good Old Days
• But
• Lack of feedback, depth of study and analysis
  (formative only coursework?)
• Seen exams?

14
Cheat sheets in exams

• US self-report data
• 1963 1995-6
• 16 17 11
• Source McCabe et al. (2001)

15
Raise detection levels

• Clues
• Electronic detection
• Phonelines
• Secret cameras in student accommodation

16
Deal with in a consistent and transparent way

• Have and publish policies
• Consistency between cases
• Consistency across disciplines
• Consistency within institutions
• Consistency across institutions
• Role for accrediting agencies and the QAA

17
Office of the Independent Adjudicator

• The greatest difficulty lies in the inconsistent
  approach of HEIs to those guilty of plagiarism,
  whether the response is the retaking of a module,
  further fees, failure, or disciplinary action.
  Some cohesion is necessary here.
• Deech (2004)

18
Publish outcomes

• Sweden - students
• USA funded researchers
• Office of Research Intergity

19
Methods in use

• All Some
• Guidance 93 7
• Assessment design 22 52
• Oral examinations 0 18
• Detection software 7 15
• Search engines 11 59
• Auditing project data 11 56

20
Selected sources

• Appleton Carroll (2001)
• Carroll (2002)
• Galles (1997)
• McCabe Pavela (1997)
• Penn State website
• And many many others