Transcript Tutorial 3

Faculty of Science and Foundation
Studies
MEASUREMENT AND
EVALUATION IN
SCIENCE
TUTORIAL THREE
PRESENTATION
Assoc. Prof. Dr Abdul Rashid Johar
SBSC4103 Measurement and Evaluation In Science
Jan 2005
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
COVERAGE OF TUTORIAL
TUTORIAL THREE COVERS:
REVISIT CHAPTERS 4
CHAPTER 5 pg 97 -128
CHAPTER 6 pg 129 -152
SUMMARY pg 128
SUMMARY pg 150
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
Constructing Objective Test Items
• True False Items
• Matching Items
• Fill-in-the blanks Items
• Completion Items
• Multiple Choice Items
• Suggestions for Writing MCQ Item
• Scoring Objective Tests
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
Constructing Subjective Test Items
• Short Answer Items
• Essay Items
• Suggestions for Writing Essay Items
• Scoring Essay Response
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
CHAPTER 5 REPORTING TEST SCORES
and ITEM ANALYSIS
Scales of Measurement
 Nominal scale
 Ordinal scale
 Interval scale
 Ratio scale
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
Nominal Scale
• Examples of nominal scale:
 Matriculation number
 Identity card number
 Plate number of cars
 House number
 Classification/categorical number such for gender,
locality, or socioeconomic status.
• Mode can be calculated and interpreted for nominal
scale. Mean and median cannot be calculated for the
scale.
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
Ordinal Scale
• Examples of ordinal scale:
 Ranking of students based on marks
 Ranking of pupils based on parents’ economic status
 Ranking of students according to height
 Ranking statements according to priority or importance
 Use of rating scales
• Mode and Median can be calculated and
interpreted for ordinal scales.
• Mean cannot be calculated for this scale. It is
meaningless.
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
Interval Scale
• Examples of interval scales:
 marks in science objective tests
 marks in essay tests
 temperature in Fahrenheit or Centigrade
 time in seconds or minutes
• Interval scale starts from zero, but this zero is
relative. This zero is not absolute.
• Mean can be calculated and interpreted for
interval scale.
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
Ratio Scale
• Examples of ratio scales:
 weight of material or matter in gram or kilogram
 volume of liquid or container
 temperature in Kelvin
 length in cm or kilometer
• Ratio scale starts from zero, but this zero is
absolute.
• Mean can be calculated and interpreted for ratio
scale.
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
Descriptive Statistics
• Rank-ordered distribution
• Ungrouped frequency distribution
• Grouped frequency distribution
• Measures of central tendency
 Mean, median and mode
• Measures of variability/dispersion
 Range, semi inter-quartile range, and
 standard deviation
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
Exercise 1
i. Calculate the mode of ungrouped frequency
distribution:
24, 25, 28, 29, 29, 29, 29, 30, 30, 30, 31, 31, 32
ii. Calculate the median of the above distribution:
iii. Calculate the mean of the above distribution.
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
Exercise 2
i. Calculate the mean, median and mode of the
following group frequency distribution:
Class Interval
f
45-49
40-44
35-39
30-34
25-29
20-24
6
12
18
13
7
4
ii. Calculate the standard deviation of the above
distribution. Find the variance.
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
Measures of Relative Position
• z-scores
Mean= 0,
Standard deviation= 1,
Range from -5 to +5
• T-scores
Mean= 50,
Std. deviation= 10,
Range from 0 to 100
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
Calculation of z scores and T scores
• For one student X in three subjects
• For three students P, Q, and R in a single subject
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
Item Analysis
Difficulty Index of Item
The more difficult an item is the higher the
difficulty index.
Discrimination Index of Item
The higher the ability to differentiate between
groups the higher the discrimination index.
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
Calculation index of difficulty
• Proportion of right answers from upper group
students.
• Proportion of right answers from lower group
students.
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
Calculation of index of discrimination
• A good item will discriminate good students and
not so good students
• Index range from -1 to 0 to +1.
• Discrimination index for a good item is equal and
above +0.4
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
CHAPTER 6 Performance-Based and
Portfolio Assessment
a. What is performance assessment?
b. The importance of performance-based
assessment.
c. How would you do the performance-based
assessment in school?
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
The Purposes of Performance
Assessment
1. Monitoring student progress towards the desired
outcomes
2. Certifying student skills and capabilities
3. Holding schools and teachers accountable for
student achievement
4. Achieving better alignment of curriculum,
instruction and assessment
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
Developing Performance Assessment
1. Deciding on what to test
 What knowledge or content?
 What intellectual skills?
 What are the important dispositions of mind to
successfully perform with the knowledge?
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
Developing Performance Assessment
2. Designing the Assessment Context.
 Create the task or situation for student to
demonstrate the knowledge or content
 Create task or project for student to use those
intellectual skills
 Create task involving debate, presentation or
experiment on issues important to the topics in
school curriculum.
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
Developing Performance Assessment
3. Specifying the Scoring Rubrics
 Performance tests require time to score them.
Projects, portfolios and practical work demand
commitment from teachers when scoring them.
 To overcome this demand, limit the scoring criteria
to those qualities of performance that are easiest
and most important to rate for doing an effective
job.
 Rubrics provide a set of guidelines that help
students to monitor their work and the teachers to
focus on teaching and assessment.
(refer to pages 141 to 147 for details)
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
Developing Performance Assessment
4. Specifying the Testing Constraints
 Projects, portfolios and practical work demand
commitment from teachers when scoring them.
Teachers should give time limits for the work to be
completed.
 Teachers should also give references and materials
to be used. The number has to be limited and how
much help from teachers and peers has to be
recorded.
 What are the appropriate or authentic limits
allowed to gain the standard required in the
assessment.
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
Closure of Tutorial 3
1. Tutors must summarize chapters 5 and 6 briefly.
2. Tutors must further encourage students to do some
exercises on their own.
3. Tutors must further encourage all students to use online discussion. Any question can be asked
concerning these chapters.
4. Remind students that they should be ready with
their ASSIGNMENT by next tutorial.
5. Remind students that Test 2 will be held in Tutorial
4.
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia
Faculty of Science and Foundation
Studies
Thank you
Subject Matter Expert/Author: Assoc. Prof. Dr Rashid Johar (OUM)
Copyright © ODL Jan 2005 Open University Malaysia