Course manual 2022/2023

Course content

An important goal of the course is to provide students with the required knowledge of statistical and probabilistic reasoning to distinguish correct from erroneous argumentation when applied to Forensic Science. Intuitive reasoning is frequently the source of serious misconceptions that all too often have lead to wrong juridical sentences. In the course, the students will see how to recognize and avoid such mistakes through formalistic analysis.

A second goal is to provide students with a basic toolbox for statistical estimation and hypothesis testing. The course is not meant as an advanced statistics course, but we will spend considerable effort on understanding and applying statistical tests such as the standard normal test, the student-t test and – ultimately - the chi-square test.

Study materials

Literature

Aitken and Taroni. Statistics and the Evaluation of Evidence for Forensic Scientists. John Wiley And Sons Ltd., Second edition, 2004. (this is the text book for this course)

Schneps and C. Colmez. Math on Trial. Ingram Publishers Services US, 2013. (we will use this book to learn examples of erroneous probabilistic and statistical reasoning in the court room)

Syllabus

There slides used in class will be available on canvas.

Practical training material

Exercises will be available on canvas.

Other

For students that have difficulty with the subject, there are may additional sources available. The following are at the correct level for this course. Note that these references will NOT be used in the course and they will not be part of the examinations either. They are meant as additional material for students that struggle with the level of the course:

Introduction to Statistics - Online Edition. David M. Lane (ed.). With contributions by David Scott, Mikki Hebl, Rudy Guerra, Dan Osherson, and Heidi Zimmer

Essential Mathematics and Statistics for Forensic Science. Craig Adam. Wiley, 2010.

Objectives

1. Use basic concepts relevant for statistical and probabilistic analysis and hypothesis testing.
2. Apply basic statistical and probabilistic methods and techniques to stylized forensic case formulations (probability, discrete and continuous distributions, hypothesis testing).
3. Evaluate the risks (i.e. possibilities of drawing wrong conclusions) and detect erroneous use of probabilistic and statistical methods and techniques when applied to e.g. data sets, excerpts of criminal trials and scientific articles.
4. Present a critical analysis of the use of statistical tools in the forensic context in a clear, coherent and understandable way.

Teaching methods

Lecture
Presentation/symposium
Self-study
Working independently on e.g. a project or thesis
Discussion on canvas forum
Perusall (tool to jointly study the text book)

Prior to the first lecture the course starts with an individual preparation/refresher about statistical concepts (mean, median, mode, variance, covariance, correlation, regression) and probabilistic fundamentals (probability space, axioms of probability, basic combinatorics, conditional probability).

In weeks 37-41 the theory classes with accompanying exercise classes will provide students with the required theoretical knowledge. There are also weekly question hours where more theoretical issues and more elaborate problems (like old exam problems) can be discussed.

In week 39 there will be a midterm test about the material covered in self-study and during the first two weeks.

In the second part of the course, the emphasis will be on the use of statistical estimation and hypothesis testing, using the same format of theory, exercise and question hour classes.

In parallel, the students will work in small teams on an assignment to critically analyze a number of criminal trials (each group is assigned a Chapter from the book Math on trial). The cases serve as an illustration of the impact erroneous reasoning may have on the course of justice. The focus in the assignments is to identify the correctness in using statistical and probabilistic analysis and techniques in the forensic practice. The different cases will give the student insight into a wide range of applications as well as a broad spectrum of erroneous reasoning in forensic applications, e.g., alleged murder, DNA analysis, database trawling and handwriting comparisons.

In week 42, the groups will present their work in a full-day meeting. The exam is in week 43.

Learning activities

Activity	Hours	Preparation
Theory Lectures	10	10
Exercises/Problems/Q&A	20	40
Midterm & Exam	5	15
Presentations	8	32
Total	140	(5 EC x 28 uur)

Attendance

This programme does not have requirements concerning attendance (OER part B).

Additional requirements for this course:

attendance (and participation) is mandatory for

- the midterm test

- the Math on Trial presentations

If a student can not be present at some of these for major reasons, if possible this must be reported to the lecturer at least one week ahead, or otherwise the lecturer needs to be contacted as soon as possible after the missed event. For the Math on Trial presentations a special arrangement (partial attendance/additional assignment) may be possible in special cases. For the midterm it will be possible to have a separate test jointly with the resit of the exam.

Assessment

Item and weight	Details
Final grade
15% Midterm test	Must be ≥ 5.5
35% Individual grade for the group assignment	Must be ≥ 5.5
50% Exam 27-10-2022	Must be ≥ 5.5

The components will be weighted as follows:

Midterm test (15%)
Group assignment including presentation (chapter from the book Math on Trial, or an advanced exam problem)
Final exam (50%)
Contributions to the discussion forum

All components will be graded on a scale from 1 to 10, with a maximum of one decimal after the point. These grades are used to calculate the final grade. In order to pass the course, each of the three components (midterm test, Math on Trial, and final exam) must be sufficient, i.e. at least a five and a half. When a student has not fulfilled this requirement, the examiner will register the mark ‘did not fulfil all requirements’ (NAV) whether or not the averaged grade is sufficient.

Resit

For the course Forensic Science there is one single resit to make up for possible insufficient grades for the midterm test and the final exam. As is usual in the program MFS, at the resit you can choose to do a resit for the final exam only, in case you already passed the midterm test (grade at least 5.5). Those who take the resit for the midterm and the final exam, must answer additional questions. It is also possible to do a resit only for the midterm in case you passed the final exam.

Change in format

Note: the formats of the midterm test, the final exam and the resit are as those of previous tests/exams since the academic year 2018-2019. Exams of earlier years are very suitable as exercise material, but differ in their formats (e.g., there was no midterm test prior to 2018 and there was no text book for use at the exams).

Attendance

Attendance is mandatory, unless there are strong reasons to miss a class. To get permission to miss a class you need to request it well in time (at least one week ahead). On September 29, we will have the midterm test in class. No exception can be made for that.

The final grade will be announced at the latest 15 working days after the final course activity (November 17th). Between this date and 35 working days after the final course activity (December 15th), a post-exam discussion or inspection moment will be planned. This will be announced on Canvas and/or via email

LO	Tested in component	EQ 1	EQ 2	EQ 3	EQ 4	EQ 5	EQ 6	EQ 7	EQ 8	EQ 9	EQ 10
1	1, 3, 4				x
2	2, 3, 4			x
3	2, 3						x
4	2									x

Table of specification: the relation between the learning outcomes of the course (see 1.3), the assessment components of the course (see 2.4) and the Exit Qualifications (EQ) of the Master’s Forensic Science (described in the Introduction in the Course Catalogue)

Students that were enrolled in the course in previous years

Students that took part in the Math on Trial assignment in the previous year need not participate in that part this year; the grade of last year is still valid.

All other grades from last year are not valid.

Inspection of assessed work

Via announcements on canvas.

Students need to contact the lecturers to make an appointment.

Assignments

The Math on Trial assignment is a group assignment. The individual grade fro this assignment can differ at most one point from the group grade based on the student's performance in Perusall, the canvas forum, and the discussions during the Math on Trial presentations.

1 Midterm test

The material of the first two theory classes and accompanying exercise classes will specifically be examined in the (individual) written midterm test in week 39.

2 Group assignment: Math on Trial

To illustrate the danger in wrongly using statistical and probabilistic reasoning, all students will study a chapter of the book Math on Trial (in groups) and prepare a joint presentation about (i) the contents of this chapter, as well as (ii) their analysis of it. Two (or three) students from the group will present their results in class. The audience is formed by students from other groups who should engage in a critical debate through questions to the presenting group.

The slide book and the presentation will be graded (one grade for the group). The individual grade of each student can deviate up to one grade point from the group’s grade, depending on their individual participation in discussions about their own presentation and that of other groups. The student’s individual grade for the assignment must not be lower than 5.5 in order to pass the course and will contribute to the final grade (see Section 2.4 for details).

3 The final examination

The material of the four theory classes and accompanying exercise classes will specifically be examined in the (individual) final written exam. The result of the exam must not be lower than 5.5 in order to pass the course (see Section 2.4 for details).

Fraud and plagiarism

The 'Regulations governing fraud and plagiarism for UvA students' applies to this course. This will be monitored carefully. Upon suspicion of fraud or plagiarism the Examinations Board of the programme will be informed. For the 'Regulations governing fraud and plagiarism for UvA students' see: www.student.uva.nl

Course structure

See course components.

Timetable

The schedule for this course is published on DataNose.

Additional information

The student must have the textbook available for use during the midterm test, the final exam and the resit. These tests/exams are open book and the student may need to consult specific sections or tables in the book.

Last year's student feedback

In order to provide students some insight how we use the feedback of student evaluations to enhance the quality of education, we decided to include the table below in all course guides.

Statistics for Forensic Science (5EC)	N 19
Strengths Q&A sessions The teachers enthusiasm, both teachers were very approachable Amount of exercises available Teacher put a lot of effort in making a digital connection with the students that had to stay at home	Notes for improvement Math on trial presentation, students felt they didn’t have enough time to prepare. Grading of the participation. Participation on the discussion board on Canvas was graded but some students were not aware of this. Sindo mentioned that their participation would be graded but it was not clear that he meant on the Canvas discussion board. It was also not clear that students would get minus points for not participating. The type of practice exams Structure of the course
Response lecturer: The assignments can be published earlier (even in the first week) - but the students can only make an educated choice after the third week. Setting the deadline for revealing their preferences earlier, means that they can only take a chance at what they are actually interested in. And the groups can only start working after all the preferences have been revealed and - based on these preferences - the assignments are allocated to groups. The lecturer fears that next year's students will not appreciate this very much. The course manual did mention that the discussion forum counted toward the individual grade, and the lecturer reminded of this several times - perhaps some students did still not know, but most surely did! The lecturer does agree that more details can be given about the numerical computation toward the grade. Meaningful participation in discussions (say one post per week) ensures that the individual grade will not be lower than the group grade. Making errors does not lead to deduction from the grade. Correct contributions on the other hand do have a positive effect on the grade. (The only way to have a lower grade than the group grade is by not participating in discussions, or insisting on errors that were identified by a lecturer.) The exam has been open book for four years now (upon repeated request by students in the past). Most practice exams were from this period - so the complaint is incorrect. We did also discuss problems from the "closed book" era which are still valuable - answers to some questions can be literally found in the book, but are excellent practicing material nonetheless (in closed book exams, these type of questions are now included in the problem description with reference to their appearance in the book). The main difference between open and closed book exams is that in an open book exam we refer to parts in the book for further context and technical details. The lecturer is unsure about this remark. If it is meant that there should be reference to a specific explanation in a theory class for each problem that we discuss, then that is virtually impossible. One of the main goals of the course is to let students combine different theoretical concepts in answering a practical question. Although our problems remain quite stylized examples of what may be true practice, they are challenging enough to let students think about which assumptions/conditions are reasonable in the stylized examples. Thus, for each problem all previously treated theory should be included.

Owner	Master Forensic Science
Coordinator	prof. dr. R. Nunez Queija
Part of	Master Forensic Science, year 1

Statistics for Forensic Science