Assessment

A. Assessment Scenarios (Validity)

Read and analyze carefully each scenario and determine what type of validity is used.
Explain your answers.

Scenario 1

Test constructors in a secondary school designed a new measurement procedure to

measure intellectual ability. Compared to well-established measures of intellectual ability, the
new test is shorter to reduce the arduous effect of a long test on students. To determine the
effectiveness, a sample of students’ accomplished two tests – a standardized intelligence test and
the new test with only a few days interval. Results from both assessments revealed high
correlation.

Construct Validity, the example scenario verified by comparing the test to other test that

measure similar qualities to see how highly correlated the two measures are. It also determines

which assessment is meaningful measure of an unobservable trait or characteristic like

intelligence, reading comprehension, honesty, motivation, attitude, learning style, and anxiety.

Scenario 2

After the review sessions, a simulated examination was given to graduating students a few
months before the Licensure Examination for Teachers (LET). When the results of the LET
came out, the review conductor found that the scores in the simulated (mock) examination are
not significantly correlated with the LET scores.

Criterion - related validity (Predictive Validity) is the extent to which a test or

questionnaire predicts some future or desired outcome. The criterion measures which the test

scores validated and obtained.. According to the given example a simulated examination was

given to graduating students a few months before the LET. So when the results of the LET came

14
out. They found out that the simulated examination aren’t significantly correlated.

Scenario 3

A new test was used as a qualifying examination for Secondary Education freshmen who
would like to major in Biological Science. The test was developed to measure students’
knowledge of Biology. The test as then administered to two groups of sophomores: those
specializing in Social Studies and those already majoring Biological Science. It was
hypothesized that the BioSci major will score better in the assessment procedure. Test results
indicated that it is so.

Criterion-related Validity (Concurrent Validity) indicates the correlation between the

predictor and criterion when data on both were collected at around the same time. The test was

developed to measure student’s knowledge of biology. They created a new test to be used as a

qualifying examination for Secondary Education freshmen. They administered two groups of

sophomores, those specializing in Social Studies and those already majoring Biological Science.

The students majoring in BioSci got better scores. Therefore, the test develop has high validity to

measures student knowledge in biology.

Scenario 4

A science teacher gave a test on volcanoes to Grade 9 students. The test indicated the
types of volcanoes, volcanic eruptions and energy from volcanoes. The teacher was only able to
cover extensively the first two topics. Several test items were included on volcanic and how
energy from volcanoes may be tapped for human use. Majority of her students got low marks.

Content Validity, the extent to which the content or topic of the test is truly representative of the

course. According to the given example, the test about volcanoes may lack content validity it’s

because the teacher only covers the first two topics but the last topic the teacher didn’t discuss

15
about it. Several test items included on volcanic and how energy from volcanoes may be tapped

for human use. So, the result of the test is that the majority of her Grade 9 students got low

marks. But if the teacher discuss all the topics to her students and her students got higher scores

then it will gain high content validity.

Scenario 5

A teacher handling “Media and Information Literacy” prepared a test on “Current and
Future Trends of Media and Information”. Topics include massive open online content, wearable
technology, 3D environment and ubiquitous learning. Below are the learning competencies:

The learners should be able to:

a. evaluate current trends in media and information and how they will affect individuals and
the society in general;
b. describe massive open online content;
c. predict future media innovation; and
d. synthesize overall knowledge about media and information skills for producing a
prototype of what the learners think is a future media innovation.

The teacher constructed a table of specification indicating the number of items for each topic.
The test items target remembering, understanding, and applying levels of cognitive domain.

Face validity refers on the judgment on the appropriateness, suitability, and mechanics in the

construction of the tests. The extent to which a test appears to measure what it is intended to

measure. A test in which most people would agree that the test items appear to measure what the

test is intended to measure would have strong face validity.

B. Assessment Scenarios (Reliability)

16
 Read and analyze carefully each scenario and determine what method or establishing
reliability is used. Explain your answers.

Scenario 1

For a sample of 150 Grade 10 students, a Science test on Living Things and their
Environment was tested for reliability by comparing the scores obtained on the odd-numbered
and even-numbered items.

Split-half method it measures the extent to which all parts of the test contribute equally to

what is being measured. This is done by comparing the results of one half of a test with the

results from other half. In the given example comparing the scores obtained on the odd-

numbered and even-numbered items. So, the method being use in this scenario is Split-half

method.

Scenario 2

Below is a table containing ratings of two teachers on the paper submitted by six Grade 9
students about their “personal mission in life”. In rating the students’ papers, a rubric was
developed.

Student Rater A Rater B Rank (A) Rank (B)

A 14 8 5 6
B 15 12 4 4
C 18 15 2 2
D 20 16 1 1
E 12 10 6 5
F 17 14 3 3
Mean = 16 Mean = 12.5 r = 0.94
SD = 2.9 SD = 3.1

Internal Consistency Method is typically a measure based on the correlations between different

items on the same test .It measures whether several items that propose to measure the same

general construct produce similar scores.

17
Scenario 3

Scores from 100 Grade 7 students were obtained from a November administration of a
test in Filipino about panghalip na panao or personal pronouns. These were compared to the
scores of the same group from a September administration of the same year.

Test-Retest Method assesses the external consistency of a test. It measures the stability of

a test over time. A typical assessment would involve giving participants the same test on two

separate occasions. If the same or similar results are obtained then external reliability is

established.

Scenario 4

Ms. Castro, a 5th-grade Social Studies teacher wanted to find out whether her first-quarter
long test was equivalent to her first-quarter test in the same subject last year. Thus, she
administered both tests to her students.

Parallel/ Alternate Form Method when the social studies teacher wants to find out the

results of her first quarter long test and her first quarter test in the same subject last year. So she

given two different versions of the same test at different times. The scores are then compared to

see if it is a reliable form of testing.

C. Computation for Validity

1. Mr. Santos would like to find out if the test questionnaire he made for his ENG 10 class
is valid or not. He obtained a copy of a valid test from his co-worker, and administered
both examinations to the same group of students. Using the data below, solve for the
concurrent validity of the text questionnaire that Mr. Santos made.

TEST (X) TEST (Y) XY X2 Y2

34 30 1020 1156 900
40 37 1480 1600 1369
35 25 875 1225 625
49 37 1813 2401 1369

18
 50 45 2250 2500 2025
38 29 1102 1444 841
37 35 1295 1369 1225
47 40 1880 2209 1660
38 35 1330 1444 1225
45 40 1800 2025 1600
413 353 14,845 17,373 12,779

Where:
n=10 Ʃxy =14,845 Ʃ x=413 Ʃy=353 x²=17,373 y²=12,779

r= n Ʃxy-(Ʃx) (Ʃy)
√ [nƩx ² -(Ʃx)²] [nƩ²y-(Ʃy)²]

r=10 (14,845)-(413) (353)

√ [10 (17,373-(413)²] [10(12,779)-(353)²]

r=148,450-145,789
√ [173,730-170,569] [127,790-124,609]

r=2,661
√ [3,161] [3,181]

r=2,661
√ 10,055,141

r=2,661
√3,170, 98

r=0.83 A 0.83 coefficient of correlation indicates that his test has a high predictive validity.

2. Ms. Corpuz administered an exam in her Science class few months ago. She wants to
determine the predictive validity of the test, using the students’ test scores and final
grades. Solve for the predictive validity of the test she made.

TEST (X) TEST (Y) XY X2 Y2

90 40 3600 8100 1600
87 48 4176 7569 2304
86 45 3870 7396 2025
88 46 4048 7744 2116
84 47 3948 7056 2209

19
 86 42 3612 7396 1764
85 47 3995 7225 2209
83 44 3625 6889 1936
89 45 4005 7921 2025
90 44 3960 8100 1936

868 448 38.866 75,396 20,124

Where:
n=10 Ʃx=868 Ʃy=448 Ʃxy=38,866 x²=75,396 x²=75, 396 y²=20,124
r=n Ʃxy-(Ʃx) (Ʃy)
√ [nƩx ² -(Ʃx)²] [nƩ²y-(Ʃy)²]
r=10 (38,866)-(868) (448)
√[10(75,396)-(868)²[10(20,124)-(448)²]
r=388,660-388,864
√[753,960-753,424] [201,240-200,704]
r= -204_____
√[536] [536]
r= -204______
√287,296
r= -204______ = -0.38 Negative coefficient correlation.
536

D. Computation for Reliability

1. Solve for the reliability of a test using the TEST-RETEST METHOD.

TEST (X) TEST (Y) XY X2 Y2

40 45 1800 1600 2025
43 42 1806 1849 1764
41 47 1927 1681 2209
39 44 1716 1521 1936
45 47 2115 2025 2209
41 44 1804 1681 1936
43 45 1935 1849 2025
42 44 1848 1764 1936
46 43 1978 2116 1849
47 45 2115 2209 2025
427 446 19,044 18,295 19,914

20
Where:

n=10 Ʃxy=19,044 Ʃx=427 Ʃy=446 x²=18,295 y²=19,914

r=n Ʃxy-(Ʃx) (Ʃy)
√ [nƩx ² -(Ʃx)²] [nƩy²-(Ʃy)²]
r=10 (19,044)-(427)(446)
√10 (18,295)-(427)²[199,140-198,916]
r=190,440- 190,442
√[182,950-182,329] [199,140-198,916]
r= -2_____
√[64] [224]
r=-2______
√[621] [224]
r=-2_____
√139,104

r=-2______
372.96
= 0.00536 or 0.005

The obtained value of 0.005 means a very low relationship thus, the results in statistics test are
not reliable.

2. Solve for the reliability of a test using the ALTERNATRE FORM METHOD.

TEST (X) TEST (Y) XY X2 Y2

45 49 2205 2025 2401
43 38 1634 1849 1444
40 43 1720 1600 1849
41 43 1763 1681 1849
39 43 1677 1521 1849
46 44 2024 2116 1936
43 47 2021 1849 2209
40 42 1680 1600 1764
38 39 1482 1444 1521

21
 46 43 1978 2116 1849
421 431 18,184 17,801 18,671
Where:
n=10 Ʃxy=18,184 Ʃx=421 Ʃy=431 x²=17,801 y²=18,671
r=n Ʃxy-(Ʃx) (Ʃy)
√ [nƩx ² -(Ʃx)²] [nƩ²y-(Ʃy)²]
r=10 (18,184)-(421)(431)
√[10 (17,801)-(421)²][10(18,671)-(431)²]
r=181,840- 181,451
√[178,010-177,241] [186,710-185,761]
r= 389_____
√[769] [949]

r=389______
√729, 781
r=389____
854.27
r=0.45

The obtained value of r is 0.45, which indicates a moderately high relationship. Therefore, the
test scores in the two forms of mathematics test are reliable.

3. Solve for the reliability of a test using the SPLIT-HALF METHOD.

Odd (X) Even (Y) XY X2 Y2

14 19 266 196 361
19 18 342 361 324
17 18 306 289 324
15 13 195 225 169
20 15 300 400 225
11 9 99 121 81
24 20 480 576 400
16 15 240 256 225
15 15 225 225 225
15 13 195 225 169
166 155 2,648 2,874 2,503

22
Where:

n=10 Ʃxy=2,648 Ʃx=166 Ʃy=155 x²=2,874 y²=2,503

r=n Ʃxy-(Ʃx) (Ʃy)
√ [nƩx ² -(Ʃx)²] [nƩy²-(Ʃy)²]
r=10 (2,648)-(166) (155)
√[10 (2,874)-(166)²][10(2,503)-(155)²]
r=26,480- 25,730
√[28,740 -27,556] [25,030-24,025]
r= _750____
√[1,184] [1,005]
r= 750____
√1,189,920
r= _750___
1,090.83
r=0.69

The reliability index of 0.69 was obtained. This means that the results of the test are reliability.

4. Solve for the reliability of a test using the KUDER-RICHARDSON METHOD.

Pupils Score (X) X-X̄ (X-X̄)2

A 43 -1.8 3.24
B 45 0.2 0.04
C 47 2.2 4.84
D 46 1.2 1.44
E 42 2.8 7.84
F 43 -1.8 3.24
G 47 2.2 4.84
H 45 0.2 0.04
I 41 -3.8 14.44
J 49 4.2 17.64
448 5.6 57.60

Where: Ʃx=448 k=50 Ʃ(x-¯x)² =57.60

¯x=Ʃx

23
 N
¯x=448__
10
¯x=44.8

SD² =Ʃ(x- ¯x) ²

N–1

SD² = 57.60
10-1

SD² = 57.60
9
SD² =6.4

KR₂₁= _k__ [1- ¯x (k- ¯x) ]

k-1 kS²

KR₂₁= 50 [1-44.8(50-44.8)]
50 - 1 50 (6.4)

KR₂₁= 50 [1-44.8(5.2)]
49 320
KR₂₁= 50 [1-232.96
49 320

KR₂₁ = 1.02 [ 1-0.73]

KR₂₁ = 1.02 [ 0.27]

KR₂₁ = 0.275 or 0.28

The reliability index of 0.28 was obtained. This means that the results of the test are reliable

24