logotip carneta reference center
  Self-assessment and Summative Assessment in E-education     english
logotip referalnog centra
 
 

Reference centers / Self-assessment and Summative Assessment in E-education / Knowledge assessment theory / Writen Examination / Types of questions

Types of written questions

 

Table 1. Overview of written questions

Types of written questions

Selected-response
(one of the offered answers
should be selected)
constructed-response
(an answer should be written)
Type Brief description Type Brief description
Multiple choice question - MCQ Only one of the offered answers is correct Essay  
Multiple response More than one answer can be correct Textual entry There are no answers offered, a student has to write the answer into an empty field
True / False The question is a statement the accuracy of which should be evaluated (only two options) Numerical entry A student writes a numerical answer  into an empty field
Matching questions Notions written in two columns should be matched Gap filling A student fills in the words missing from the text

 SELECTED-RESPONSE QUESTIONS

 Multiple choice question - MCQ

    Advantages:

- jednostavno, precizno i objektivno ocjenjivanje/bodovanje; :
- this assessment/grading system is simple, precise and objective; it is especially convenient for computer awarded point system - it significantly reduces and simplifies professor's job with grading
- convenient for self-assessment, as well as for summative exams
- it is possible to examine different cognitive levels of knowledge - according to Bloom's taxonomy of educational goals: acquiring facts, understanding, application of factual knowledge, analysis, synthesis, evaluation)
- quick and simple statistical analysis of the whole test, as well as of the difficulty of individual questions and discriminative ability of a question ("item analysis")
- MCQ tests examine the subject content more comprehensively than essay type questions.
- simple composition and analysis of MCQ tests enables frequent testing, so professors (and students) get regular feedback about students' success in mastering the matter
- in MCQ tests students don't have the possibility to skirt or simplify the theme (unlike essay type tests) - the professor can determine in detail the plan ("depth") of the exam

    Problems:
- writing good quality questions is complex and time-consuming
- professors are prone to writing questions which examine only the memorizing of facts since that type of questions is the easiest to create
- these tests often don't measure the real knowledge since the right answer can be deducted by eliminating the incorrect ones
- a certain percentage of answers can be guessed
- it is very difficult to test creativity with this type of questions (it is tested best through essay type questions)

 Multiple response questions

- their advantage is easy automated result analysis 
- their downside is usually low level of knowledge which is being tested
- and a greater possibility of guessing the right answers

 True / Fals question)

- subcategory of MCQ tests with only two choices - "true" or "false"
- a question is written in the form of a statement which needs to be evaluated as true or false
- this type of questions is suitable for short formative self-assessment which gives a student (or a professor) quick feedback on the level of acquired factual knowledge
- suitable for factual knowledge examination
- not suitable for examination of higher cognitive knowledge (see Bloom taxonomy)
- the statement should be written using different vocabulary from the one used in the textbook or instructional materials, so that simply memorizing the text will not be sufficient to select the right answer
- when writing the statement, words that could suggest what the true/false answer is, should be avoided; e.g. words such as "never", "nothing", "always", "everything" have a tendency towards false, while adverbs such as "usually", "generally", "sometimes" or "often" are probably true
- it is desirable to avoid negative statements
- it is recommended to have approximately the same number of true and false statements in a test
- statistic probability to guess the right answer without actual knowledge (randomly) is 50%, so the minimum test score percentage should be at least 75% (with MCQ with 5 options the minimum test score is 60%)

 Matching list

- subcategory of MCQ where notions in two columns must be mutually matched
- these questions are suitable for both formative self-assessment through which students and professors get quick feedback about the level of acquired knowledge, and for summative exams
- it enables examination of factual knowledge, and adept question construction can enable examination of higher cognitive levels of knowledge such as causal connection (see Bloom's taxonomy)
- it is important that information in both columns are as homogenous as possible
- reusability of answers is important
- it is recommended to have more possible answers than questions
 

 CONSTRUCTED-RESPONSE QUESTIONS

     Text match

- constructed-response questions are formed similarly as MCQ, but without the offered answer. In stead, students have to write the answer into free space provided for it
- unlike MCQ where students choose between offered answers (there is a possibility that they do not have the necessary knowledge, but choose the correct answer by elimination or randomly), with this type of questions the students must know the correct answer
- if the problem is numerical and the answer is a number, all possible variants which are considered correct should be predicted (e.g. decimal places)

    Fill in the blank

- advantage: it is easy to examine higher cognitive levels
- downside: harder automated analysis, since all the forms of the correct answer have to be predicted

    Essay

        There are two types of essays:
- long, extended essay which is used to examine one sample of subject matter on a higher cognitive knowledge level (creativity)
- short, limited essay - several short essays can be used for a more superficial processing of a larger number of samples of subject matter
       Advantages:
- it is the only examination method which can be used to assess the students' ability to explain a certain answer independently and in a written form
- higher cognitive levels of knowledge are examined (analysis, synthesis, evaluation)
- relatively simple development and formulation of questions
      Problems:
- time-consuming analysis and grading
- potential unreliability and subjectivity in grading
- questionable validity and reliability
        Recommendations:
- use only to examine the types of activities which cannot be successfully assessed through another type of examination
- select the task suitable for instructional goals and which is the representative sample of the subject matter
- guide the students with clear instructions (e.g. explain the mechanism through which A influences B; analyse relations between A and B)

 

 
 

Copyright © CARNet
design: logotip kimajaka