Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Vad bör en forensisk expert inte uttala sig om?

Petersson, Marcus LU (2025) JURM02 20252
Department of Law
Faculty of Law
Abstract
My purpose with this thesis is to examine how well forensic experts` understanding of what they should and should not express opinions on is in accordance with an idealized view of the expert role. Given my purpose with the thesis, eleven statements were presented in a survey to 50 forensic experts working at the Swedish National Forensic Centre (NFC). Two deviations from the ideal view of an expert are tested in the survey. A forensic expert can deviate from the ideal view of an expert by assessing the probability of the proposition being true instead of how strongly the evidence supports the hypothesis. Further, a forensic expert can deviate from the ideal view by confusing the support for a proposition on source level with support for a... (More)
My purpose with this thesis is to examine how well forensic experts` understanding of what they should and should not express opinions on is in accordance with an idealized view of the expert role. Given my purpose with the thesis, eleven statements were presented in a survey to 50 forensic experts working at the Swedish National Forensic Centre (NFC). Two deviations from the ideal view of an expert are tested in the survey. A forensic expert can deviate from the ideal view of an expert by assessing the probability of the proposition being true instead of how strongly the evidence supports the hypothesis. Further, a forensic expert can deviate from the ideal view by confusing the support for a proposition on source level with support for a proposition on activity level. The survey consists of eight incorrect statements and three correct statements, covering multiple forensic areas.

The ideal view of a forensic expert acts as a starting point for the analysis of the survey responses. In addition, the thesis consists of a description of how forensic experts at NFC work, as well as how they are educated. The survey was approved by the management at NFC and NFC will gain valuable information from the survey results.

The survey results were compiled in tables and graphs. Two of the incorrect statements entail the fallacy of assessing the probability of the proposition being true, given the evidence. The percentages of the respondents that identified these statements as incorrect were 68 % and 54 %, respectively. The percentages of respondents who correctly explained why the statements were incorrect were 56 % and 22 %, respectively.

Two other incorrect statements test the forensic experts` ability to determine whether assessments of the probability that the proposition is true should be made. These assessments are also understood as decisions to treat the proposition as true. The percentages of the respondents that identified these statements as incorrect were 58 % and 64 %, respectively. The percentages of the respondents who correctly explained why the statements were incorrect were 17 % and 19 %, respectively.

The remaining four incorrect statements confuse the support for a proposition on source level with support for a proposition on activity level The percentages of the respondents that identified these statements as incorrect were 80 %, 100 %, 84% and 96 %, respectively. The percentages of respondents who correctly explained why the statements were incorrect were 68 %, 82 %, 60 % and 65 %, respectively.

The total result for the deviation of assessing the probability of the proposition being true shows that 61 % of the forensic experts at NFC correctly identified the statements as incorrect. The percentages of respondents who correctly explained why the statements were incorrect were 30 %. A lower level of accuracy is observed among the respondents compared to the deviation of confusing the support for a hypothesis on source level with support for a hypothesis on activity level. The total result for the deviation of confusing the support for a hypothesis on source level with support for a hypothesis on activity level show that 87 % of the forensic experts at NFC. The percentages of respondents who correctly explained why the statements were incorrect were 64 %. A higher level of accuracy is observed among the respondents compared to the first deviation tested.

A significance test is conducted for the results of each statement to determine whether the proportion of yes responses is higher than the proportion of no responses. The proportion of yes responses is significantly lower or higher than the proportion of no responses for six out of eleven statements. For the remaining five statements, the test is either not reliable or served no purpose. It is not possible to determine dependence between different variable for the proportions of yes answers and no answers in relation to forensic area or length of service. (Less)
Abstract (Swedish)
Mitt syfte med uppsatsen är att undersöka hur väl forensikers egen uppfattning om vad de bör uttala sig om och vad de inte bör uttala sig om stämmer överens med en idealbild av expertrollen. Mot bakgrund av syftet ställdes elva påståenden till 50 forensiska experter vid Nationellt forensiskt centrum (NFC) i form av en enkätundersökning. De sätt på vilka forensiker kan avvika från idealet som testas är uttalande som innebär bedömningar av sannolikheten för att bevistemat är sant istället för hur starkt stöd bevistemat får av de observationer som forensikern gjort samt sammanblandning av stöd för ett bevistema på källnivå med ett bevistema på aktivitetsnivå. Enkäten utgörs av åtta felaktiga påståenden och tre korrekta påståenden och omfattar... (More)
Mitt syfte med uppsatsen är att undersöka hur väl forensikers egen uppfattning om vad de bör uttala sig om och vad de inte bör uttala sig om stämmer överens med en idealbild av expertrollen. Mot bakgrund av syftet ställdes elva påståenden till 50 forensiska experter vid Nationellt forensiskt centrum (NFC) i form av en enkätundersökning. De sätt på vilka forensiker kan avvika från idealet som testas är uttalande som innebär bedömningar av sannolikheten för att bevistemat är sant istället för hur starkt stöd bevistemat får av de observationer som forensikern gjort samt sammanblandning av stöd för ett bevistema på källnivå med ett bevistema på aktivitetsnivå. Enkäten utgörs av åtta felaktiga påståenden och tre korrekta påståenden och omfattar flera olika forensiska områden.

En idealbild av en forensisk expert verkar som en utgångpunkt för analysen av enkätsvaren. Därutöver undersöks hur forensiker vid NFC arbetar med resultatvärdering samt, på en övergripande nivå, hur utbildningen av en forensiker är utformad. Enkätundersökningen var sanktionerad av enhetsledningarna vid NFC och NFC hade kvalitetsmässigt värde av att den genomfördes.

Enkätresultaten sammanställs i tabeller och grafer. Två av de inkorrekta påståendena avser sannolikhetsbedömningar för att bevistemat är sant. Andelen forensiker som korrekt bedömde uttalandena som felaktiga var 68 % respektive 54 %. Andelen som kunde lämna en korrekt förklaring till varför en forensisk expert inte bör göra ett uttalande om sannolikheten för att bevistemat är sant var 56 % respektive 22 %.

Ytterligare två av de felaktiga påståendena testar forensikernas förmåga att avgöra om uttalanden om sannolikheten för att bevistemat är sant bör göras. Uttalandena tolkas även som beslut att betrakta bevistemat som sant. Andelen som korrekt bedömde uttalandena som felaktiga var 58 % respektive 64 %. Andelen som kunde förklara varför ett sådant uttalande inte bör göras av en forensisk expert var 17 % respektive 19 %.

De resterande fyra inkorrekta påståendena består av uttalanden som innebär sammanblandning av stöd för ett bevistema på källnivå med ett bevistema på aktivitetsnivå. Andelen som korrekt bedömde uttalandena som felaktiga var 80 %, 100 %, 84 % respektive 96 %. Andelen som kunde förklara varför ett uttalande om en aktivitet inte bör göras av en forensisk expert var 68 %, 82 %, 60 % respektive 65 %.

Totalresultatet för avvikelsen att göra ett uttalande om sannolikheten för att bevistemat är sant visar att 61 % av de forensiska experterna korrekt bedömde uttalandena som felaktiga. Andelen som kunde lämna en korrekt förklaring till att en forensisk expert inte bör göra ett uttalande om sannolikheten för att bevistemat är sant var 30 %. Resultaten visar att en lägre träffsäkerhet kunde observeras vid analys av dessa uttalanden jämfört med avvikelsen att sammanblanda stöd för ett bevistema på källnivå med stöd för ett bevistema på aktivitetsnivå. Totalresultatet för avvikelsen att göra ett uttalande som utgör en sammanblandning av stöd för ett bevistema på källnivå med stöd för ett bevistema på aktivitetsnivå visar att 87 % av de forensiska experterna vid NFC korrekt bedömde uttalandena som felaktiga. Andelen som kunde lämna en korrekt förklaring till att en forensisk expert inte bör göra ett uttalande om sannolikheten för att bevistemat är sant är 64 %. Resultaten visar att en högre träffsäkerhet kunde observeras vid analys av dessa uttalanden jämfört med den första avvikelsen.

Ett signifikanstest av resultaten för varje påstående visar att andelen ja-svar är signifikant lägre eller högre än andelen nej-svar för sex av elva påståenden. För de resterande fem påståendena är testet antingen ej tillförlitligt eller saknar mening att utföra. Det går inte att påvisa något beroende mellan olika variabler för andelen ja-svar och nej-svar mot forensiskt område respektive tjänstetid. (Less)
Please use this url to cite or link to this publication:
author
Petersson, Marcus LU
supervisor
organization
alternative title
What should a forensic expert not express opinions on?
course
JURM02 20252
year
type
H3 - Professional qualifications (4 Years - )
subject
keywords
allmän rättslära, bevisvärdering, expert, forensisk expert, expertvittne, sakkunnig, enkätstudie, Nationellt forensiskt centrum, NFC
language
Swedish
id
9216782
date added to LUP
2026-01-21 10:52:35
date last changed
2026-01-21 10:52:35
@misc{9216782,
  abstract     = {{My purpose with this thesis is to examine how well forensic experts` understanding of what they should and should not express opinions on is in accordance with an idealized view of the expert role. Given my purpose with the thesis, eleven statements were presented in a survey to 50 forensic experts working at the Swedish National Forensic Centre (NFC). Two deviations from the ideal view of an expert are tested in the survey. A forensic expert can deviate from the ideal view of an expert by assessing the probability of the proposition being true instead of how strongly the evidence supports the hypothesis. Further, a forensic expert can deviate from the ideal view by confusing the support for a proposition on source level with support for a proposition on activity level. The survey consists of eight incorrect statements and three correct statements, covering multiple forensic areas.

The ideal view of a forensic expert acts as a starting point for the analysis of the survey responses. In addition, the thesis consists of a description of how forensic experts at NFC work, as well as how they are educated. The survey was approved by the management at NFC and NFC will gain valuable information from the survey results.
 
The survey results were compiled in tables and graphs. Two of the incorrect statements entail the fallacy of assessing the probability of the proposition being true, given the evidence. The percentages of the respondents that identified these statements as incorrect were 68 % and 54 %, respectively. The percentages of respondents who correctly explained why the statements were incorrect were 56 % and 22 %, respectively. 

Two other incorrect statements test the forensic experts` ability to determine whether assessments of the probability that the proposition is true should be made. These assessments are also understood as decisions to treat the proposition as true. The percentages of the respondents that identified these statements as incorrect were 58 % and 64 %, respectively. The percentages of the respondents who correctly explained why the statements were incorrect were 17 % and 19 %, respectively.

The remaining four incorrect statements confuse the support for a proposition on source level with support for a proposition on activity level The percentages of the respondents that identified these statements as incorrect were 80 %, 100 %, 84% and 96 %, respectively. The percentages of respondents who correctly explained why the statements were incorrect were 68 %, 82 %, 60 % and 65 %, respectively.

The total result for the deviation of assessing the probability of the proposition being true shows that 61 % of the forensic experts at NFC correctly identified the statements as incorrect. The percentages of respondents who correctly explained why the statements were incorrect were 30 %. A lower level of accuracy is observed among the respondents compared to the deviation of confusing the support for a hypothesis on source level with support for a hypothesis on activity level. The total result for the deviation of confusing the support for a hypothesis on source level with support for a hypothesis on activity level show that 87 % of the forensic experts at NFC. The percentages of respondents who correctly explained why the statements were incorrect were 64 %. A higher level of accuracy is observed among the respondents compared to the first deviation tested.

A significance test is conducted for the results of each statement to determine whether the proportion of yes responses is higher than the proportion of no responses. The proportion of yes responses is significantly lower or higher than the proportion of no responses for six out of eleven statements. For the remaining five statements, the test is either not reliable or served no purpose. It is not possible to determine dependence between different variable for the proportions of yes answers and no answers in relation to forensic area or length of service.}},
  author       = {{Petersson, Marcus}},
  language     = {{swe}},
  note         = {{Student Paper}},
  title        = {{Vad bör en forensisk expert inte uttala sig om?}},
  year         = {{2025}},
}