Assessment of Learning Steps in Preparing a Test Draft
Chapter 5 – TOS 1. Using the TOS, formulate your test items.
A. Choosing the coverage of the Test 2. Prepare your exam following the different
Exam Guidelines.
1. Choose
a. Grade Level (Grade 7 to Grade 12), 3. Make a General Instructions and other
b. Grading Period (Quarter). Necessary Heading.
2. Determine the Topics included in the quarter you
4. Write an Instruction or Direction for each
chose based on the Curriculum Guide set by the
Type of Test.
Department of Education.
3. Look for Reference Materials to be used that 5. Numbering in the Exam must be continuous.
discuss the topics covered in the Quarter. You may
use these as your bases in crafting your Teacher- 6. Prepare a Key to Correction and for items 1
Made Exam. through 45 and a Scoring Rubric for items 46
to 50.
B. Steps in Preparing a Table of Specifications (TOS)
7. Proofread your Exam. Use Arial 11 and long-
1. Determine the Coverage (Content) of the Test sized bond paper.
(Based on DepEd curriculum).
Chapter 6 - Item Analysis
2. Approximate the number of minutes to be spent for
each Topic (Content). Set the total number of contact Steps in Test Preparation:
hours into 20 hours (1,200 minutes) for the entire 1. Construct a Table of Specifications (TOS)
Quarter. 2. Prepare a Test Draft.
3. Set the Total Number of Items to fifty (50). 3. Subject the draft into Item Analysis and Validation
a. 25 items Multiple Choice ( 1 to 25) 3.1. Try-Out Phase: Try out the draft test to a group
of students of similar characteristics as the intended
b. 5 items True or False ( 26 to 30) test takers.
3.2. Item Analysis Phase: From the try-out group,
c. 5 items Matching Type ( 31 to 35)
each item will be analyzed in terms of its availability
d. Short Answer ( Identification 36 to 40) and to discriminate between those who know and those
( Completion ( 41 to 45) who do not know and also its level of difficulty.
3.3. Item Revision Phase: The item analysis will
e. Essay ( 46 to 50)
provide information that will allow the teacher to
When time spent is encoded, the column for decide whether to revise or replace an item.
Computed will automatically be filled out. If the value
3.4.Validation Phase: The final draft is subjected to
is a whole number, just copy the values to the
validation, if the intent is to make use of the test as a
Adjusted column. However, if the computed value is
standard test for the particular unit or grading period.
not a whole number, round off the value to a whole
number. Two Important Characteristics of an Item:
1. Item Difficulty
Note: The Computed and Adjusted Values indicate the
2. Discrimination Index
number of items for each Topic.
Item difficulty – is the number of students who are
4. Identify the number of items per Cognitive
able to answer the item correctly divided by the total
Processes and per Topic.
number of students.
5. Identify the Item Placement. Consider the Type of
DI is equal to c/t
Test and the Cognitive Process that each Type of Test
could measure. E.g. Essay Test could measure HOTS Where DI is the item difficulty, c is the number of
while Identification could only measure LOTS. students who answered correctly and t is the total
number of students who took the exam
Note: Item Placement should follow the sequence
from Left to Right and Top to Bottom.
refers to the center of a distribution of
observations.
A. Mean
B. Median
Discrimination Index - a measure that tells whether C. Mode
an item can discriminate between those who know
and those who do not know the answer. It is the Mean – is the most popular and most well known
difference between the proportion of the top scorers measure of central tendency. It is equal to the sum of
who got an item correct and the proportion of the all the values in the data set divided by the number of
lowest scorers who got the item right. values in the data set.
Formula: Dx = DIU- DIL Median - is the middle score for a set of scores
arranged from lowest to highest. Is the most
where Dx = Discrimination Index appropriate measure when there are outliers.
DIU = Index of Difficulty with respect to the Upper 25%
DIL = Index of Difficulty with respect to the Lower 25%
Discrimination Index - is the ability of an item to
differentiate among students on the basis of how well
they know the material being tested. Mode is the most frequently occurring score in the
data set. On a histogram or bar chart it represents the
A more sophisticated way of computing the item
highest bar.
discrimination index provided by ScorePak® is a
Pearson Product Moment Correlation between 2 modes = bimodal distribution
student responses to a particular item and total
A score distribution a sample has a Normal
scores on all other items on the test.
Distribution
This is equivalent to a point-biserial coefficient. It
when most of the values are aggregated
provides an estimate of the degree to which an
around the mean, and the number of values
individual item is measuring the same thing as the rest
decrease as you move below or above the
of the items.
mean
Since the discrimination index reflects the degree to the bar graph of frequencies of a normally
which an item and the rest as a whole are measuring a distributed sample will look like a bell curve.
unitary ability or attribute, the mean is equal to the median and median
is equal to the mode
values of the coefficient will tend to be lower
for tests measuring a wide range of content
areas than for more homogeneous tests.
Items with low discrimination indices are
often unambiguously worded and should be
examined.
A negative value may indicate that the item is
miskeyed.
CHAPTER 7
Measures of Central Tendency and
Dispersion/Variability
A measure of central tendency A score distribution is a Negatively Skewed
is a single value that attempts to describe a Distribution
set of data (like scores) by identifying the
central position within that set of data or if the mean is less than the median and the
scores. mode.
is also called measures of central position. the scores tend to congregate at the upper
end of the score distribution.
if scores tend to be high because teacher Then students’ score in the assessment phase of the
taught very well and students are highly lesson will tend to congregate on the lower end of the
motivated to learn. score distribution. Score distribution will be positively
skewed.
Measures of Variability - indicate how spread out a
group of scores is or how varied the scores are. The
terms variability, spread and dispersion are synonyms.
A. Range
B. Variance
C. Standard Deviation
Range - is the most simple measure of variability. It is
A score distribution is a Positively Skewed simply the highest score minus the lowest score.
Distribution Variance – is the average of the squared difference of
if the mean is greater than the median and the scores from the mean.
the mode.
the scores tend to congregate at the lower
end of the score distribution.
if scores tend to be low when the teacher
does not teach well and students are poorly Standard deviation – is a measure of dispersion, the
motivated. more dispersed the data, the less consistent the data
are.
CHAPTER 8
Norm-referenced Grading - refers to a grading system
where a student’s grade is placed in relation to the
performance of a group. It is based on a pre-
established formula regarding the percentage or ratio
of students within a whole class who will be assigned
each grade or mark.
Problem of Equivalency - ‘unfair” system of grading
when the same raw score would get a different
equivalent grade depending on the raw scores of
If teachers other students in the group.
a. teach in accordance with the principles of Uses of Norm-referenced Grading:
outcome-based teaching-learning Screening selected student populations in conditions
b. align content and assessment with the where it is known that not all students can advance
intended learning outcomes due to limitations such as available places, jobs or
c. re-teach till mastery what has not been other controlling factors.
understood as revealed by the formative
assessment Screening selected student populations in conditions
Then students’ score in the assessment phase of the where it is known that not all students can advance
lesson will tend to congregate on the higher end of due to limitations such as available places, jobs or
the score distribution. Score distribution will be other controlling factors.
negatively skewed.
Criterion-referenced Grading - is based on a fixed
If what teachers criterion measure. Ther e is a fixed target and the
students must achieve that target in order to obtain a
teach and assess are not aligned with the intended passing grade in a course regardless of how the other
learning outcomes
students in the class perform. The scale does not Cumulative Grading System - the grade of the student
change regardless of the quality, or lack thereof, of in a grading period equals his current grading period
the students. which is assumed to have the cumulative effects of
the previous grading periods.
When is Criterion-referenced Grading System Used?
Policy Guidelines on Classroom Assessment for the K
1. The teachers agreed on the meaning of a
to 12 Basic Education, DepEd Order No. 8, s. 2015
“standard of performance” in a subject but
the quality of the students is unknown or Highlights of the new K to 12 Grading System
uneven. implemented since SY 2015-2016:
2. The work involves student collaboration or
Weights of the Components for the Different Grade
teamwork, and where there is no external
Levels and Subjects
driving factor such as needing to
systematically reduce a pool of eligible 3 Components:
students. 1. Written Work
2. Performance Tasks
Four questions in Grading
3. Quarterly Assessment
Posed by Marinila D. Svinicki (2007) of the Center for
3 Subject Clusters (Grades 1 to 10):
Teaching Effectiveness of the University of Texas in
Austin: 1. Languages, Araling Panlipunan, and
Edukasyon sa Pagpapahalaga
1. Should grades reflect absolute achievement
2. Science and Mathematics
level or achievement relative to others in the
3. Music, Arts, Physical Education and Health
same class?
(MAPEH)
2. Should grades reflect achievement only or
3 Subject Clusters (Senior High School):
nonacademic components such as attitude,
speed and diligence? 1. Core Subjects
2. All Other Subjects (Applied and Specialization)
3. Should grades report status achieved or
and Work Immersion of the Academic Track
amount of growth?
3. All Other Subjects (Applied and Specialization)
4. How can several grades on diverse skills
Grade computation
combine to give a single mark?
The work involves student collaboration or teamwork,
Test Standardization - is a process by which teacher
and where there is no external driving factor such as
or researcher-made tests are validated and item
needing to systematically reduce a pool of eligible
analyzed.
students.
Standardized Tests - are psychometric instruments
A final grade of 75 and higher in all learning areas
whose scoring systems are developed by norming the
allows the student to be promoted to the next grade
test using national samples of test-takers, centering
level.
the scoring formula to assure that the likely score
distribution describes a normal curve when graphed, Chapter 4
and then using the resulting scoring system uniformly
in a manner resembling a criterion-referenced ORGANIZATION, UTILIZATION, AND
approach. COMMUNICATION OF TEST RESULTS
Cumulative and Averaging Systems of Grading Mean of ungrouped data
Averaging System - the grade of a student on a
particular grading period equals the average of the
grades obtained in the prior grading periods and the
current grading period.
Weighted Mean - When data values are assigned 4. more consistent, and
different weights, we can compute a weighted mean. 5. more uniformly distributed
The formula below is used to compute the weighted
The distance between each score in a distribution and
mean.
the mean of that distribution (𝑥−𝑥 ̅ ) is called the
Mean of grouped data deviation score.
The mean of the deviation scores is called the
standard deviation, abbreviated often as SD. The SD
describes the mean distance of the scores around the
distribution mean.
Squaring the SD gives us another index of variability,
called the variance.
Median for Ungrouped Data. To find the median, first
sort the values (arrange them in order), then follow
one of these two procedures:
1. If the number of data values is odd, the median is
the number located in the exact middle of the list.
2. If the number of data values is even, the median is
found by computing the mean of the two middle
numbers. or you may use the formula, nth order=
n+1 Coefficient of variation is the ratio of the standard
2 deviation to the mean. It is used to compare the
Median for grouped data variability of two or more sets of data even when they
are expressed in different unit of measurement.
The formula for the coefficient of variation is shown
below:
s
cv = ∙ 100 % , sample
x
σ
cv = ∙ 100 % , population
μ
A. QUARTILES
These are three values that divide the set of scores
into four equal parts, with one-fourth of the data in
each part.
1. First Quartile (Q1)- about 25% of the data
falls at or below the first quartile
2. Second Quartile (Q2)- about 50% of the data
falls at or below the second quartile. The
median is the second quartile.
3. Third Quartile (Q3)- about 75% of the data
falls at or below the third quartile
A small measure of variability would indicate that the
data are:
1. clustered closely around the mean,
2. more homogeneous,
3. less variable,
FORMULAS: The normal distribution is a special kind of
symmetrical distribution that is most frequently used
to compare scores.
Properties of the Normal Distribution:
1. The mean, median, and more are all equal
INTERQUARTILE RANGE. It is used as a measure of the 2. The curve is symmetrical. As such, the value is
spread of data computed by getting the difference a specific area on the left is equal to the value
between the third and first quartile. of its corresponding area on the right.
SEMI-INTERQUARTILE RANGE OR QUARTILE 3. The curve changes from concave to convex
DEVIATION. It is used as a measure of the spread of and approaches the X- axis, but the tails do
data computed by getting average of the difference not touch the horizontal axis.
between the third and first quartile.
4. The total area on the curve is equal to 1.
DECILES.
Empirical Rule or the (68-95-99 Rule):
These are nine values that divide the set of scores into
ten equal parts, with one-tenth or 10% of the data in 1. 68% of the scores fall between one standard
each part. deviations below and above the mean
First Decile (D1)- about 10% of the distribution are 2. 95% of the scores fall between two standards
equal or less than decile 1 deviations below and above the mean
Second Decile (D2)-about 20% of the distribution are 3. 99.77% of the scores fall between three
equal or less than decile 2 standards deviations below and above the
Third Decile (D3)- about 30% of the distribution are mean
equal or less than decile 3
Fourth Decile (D4)- about 40% of the distribution are
equal or less than decile 4
Fifth Decile (D5)- about 50% of the distribution are
equal or less than decile 5
Sixth Decile (D6)- about 6 0% of the distribution are
equal or less than decile 6
Seventh Decile (D7)- about 70% of the distribution are
equal or less than decile 7
Eighth Decile (D8)-about 80% of the distribution are
equal or less than decile 8
Ninth Decile(D9)- about 90% of the distribution are
equal or less than decile 9 A standard score is the number of standard deviations
i by which the value of a raw score is above or below
Formula: D i=( N +1 ) , Decile for Ungrouped Data the mean value of what is being observed or
10
measured.
PERCENTILES - these are values that divide the set of
scores into one hundred equal parts, with 1% of the It is a score that has been transformed to fit a normal
data in each part. It is a number where a certain curve.
percentage of scores fall below that number. 1. z-score
FORMULA: 2. T-score
i 3. Stanine scores
Pi=( N +1 ) , Percentile for Ungrouped Data
100
A. z-score. It is a standard score, used to express
The Normal Distribution a raw score in relation to the mean and
standard deviation. This relationship is
expressed in the formula: where
z= z-score
x = mean
x = mean
s = standard deviation
T-score. It is a transformed standard score whereby a
mean of zero in a z-score is transformed into a mean
of 50. This relationship is expressed in the formula
where: z= z-score
Stanine Score. A stanine (“standard nine”) score is a
way to scale scores on a nine-point scale. It can be
used to convert any test score to a single-digit score.
Like z-scores and T-scores, a stanine score is simply a
way of indicating the position of a person in a group
relative to the other people (or things) that have been
measured.
Formula:
Stanine = 2z + 5
Chapter 8 continuation…..
