test construction and evaluation lizamarie campoamor-olegario

Test Construction and EvaluationLizamarie Campoamor-Olegario

Lizamarie Campoamor-Olegario

Test or Scale Development Procedure• Step 1. Definition of Construct• Step 2. Table of Test Specifications• Step 3. Item Writing• Step 4.

Content Validation including Factor Analysis

• Step 5. Pilot Testing to Equivalent Group

• Step 6. Item-analysis, reliability analysis

• Step 7. Final AdministrationLizamarie Campoamor-Olegario

DEFINITION OF THE CONSTRUCT

Step 1.


Step 1. Definition of Construct

• Literature review

• Scientifically defined constructs

• Popular definition

• Syllabus

backLizamarie Campoamor-Olegario

TABLE OF TEST SPECIFICATIONS

Step 2.


Step 2. Table of Test Specifications

• Table of Specifications• Subject: _______________• Grade/ Year Level: _____• Type of Test: Cognitive

(Knowledge), Psychomotor (Skills), or Affective (Attitudes)?

• Testing Time: _____ • Type of Item/s: Likert? Or rating? • Number of items: ____


Knowledge (Cognitive)

• (Source: Bloom, B., Taxonomy of Educational Objectives, 1956)


Levels of Cognitive Domain (Knowledge)• Knowledge: The remembering of previously learned

material (recall of facts)• Comprehension: The ability to grasp the meaning of

the knowledge being learned• Application: The ability to use learning materials in a

new way• Analysis: The ability to break material down into its

parts so that its organizational structure may be understood

• Synthesis: The ability to combine previous experiences with new material to form a whole new structure

• Evaluation: The ability to judge the value of material for a given purpose


Knowledge – Can students RECALL information?• Key Questions: Who, What, When, Where,

How, How much, Describe, Define, Memorize, Literal questions, Which one, Name, Label, List, Reproduce, Recall

• Example of Knowledge QuestionWhich of the following are raw materials or photosynthesis? a) Water, heat, sunlight

b) Carbon dioxide, sunlight oxygen

c) Water, carbon dioxide, sunlight

d) Sunlight, oxygen, carbohydrates

e) Water, carbon dioxide, carbohydrates


Comprehension – Can students EXPLAIN ideas?• Key Questions: Explain, Describe in your

own words, Inferential questions, Summarize, What would go better, Select the definition, Read the graph table, This represents¸ Condense this paragraph, What part doesn’t fit, What are they saying, Explain what is happening, Give an example, State in 5 words, Explain what is meant, What restriction would you add, Translate, Outline, Locate, Match


Comprehension – Can students EXPLAIN ideas?• Example of Comprehension Question• If living cells similar to those found on

earth were found on another planet where there was no molecular oxygen, which cell part would most likely be absent? a)Cell membrane

b)Nucleus

c) Mitochondria

d)Ribosome

e)Chromosomes Lizamarie Campoamor-Olegario

Application – Can students USE ideas?• Key Questions: What is this used

for?, How would you use, Make a model, Tell what would happen, If…how, Demonstrate how, Construct how, Show how, How much would there be if…, Design a lesson, Choose the statements that don’t apply


Application – Can students USE ideas?• Example of Application Question• Phenylketonuria (PKU) is an autosomal recessive

condition. About one in every fifty Individuals is heterozygous for the gene but shows no symptoms of the disorder. If you select a symptom-free male and a symptom –free female at random, what is the probability that they would have a child afflicted with PKU? – (.02)(.02)(.25) = 0.0001 = 0.01%, or about 1/10,000

– (.02)(.02) = 0.0004 = 0.04%, or about 1/2,500

– (1)(50)(0) = 100% = all

– (1)(50)(0) = 0 = none

– 1/50 = 2%, or 2/100


Analysis – Do students SEE relationships?• Whole into parts, Analyse, Research, Survey,

Group, Categorise, Compare and Contrast, What inconsistencies, fallacies, Arrange, What is the relationship, Chart, What is the function of, Diagram, What conclusions, Reason for…, What does the author believe, Investigate, Make a distinction, Cause for, What motive is there, Conclude, State the point of view, Separate, What relationship, Similar, Graph, Like, Differentiate, Dissect, Categorize, Distinguish fact from fiction, fact and inference, fact from opinion, advantage from disadvantage, good from poor reason, What persuasive technique


Analysis – Do students SEE relationships?• Example of Analysis Question

Mitochondria are called the powerhouses of the cell because they make energy available for cellular metabolism. Which of the following observations is most cogent in supporting this concept of mitochondrial function? a) ATP occurs in the mitochondria

b) Mitochondria have a double membrane

c) The enzymes of the Krebs cycle, and molecules required for terminal respiration, are found n mitochondria

d) Mitochondria are found in almost all kinds of plant and animal cells

e) Mitochondria abound in muscle tissue


Synthesis – Can students combine ideas and CREATE a new entity?• Key Questions: New ways of doing,

Take risks, Consider the unexpected, Pose an alternative, Hypothesis, Create, Compose, Solve, Design, Blend, Construct, How else would you, Build, Combine, Solve the following, Imagine, Plan, Predict, Link concepts in an unusual and flexible way, Make, What if, Make a film, Invent, Propose an alternative


Evaluation – Can students make JUDGEMENTS and support them?• Evaluate quality, relevance, reliability, truth,

Which is best, Accuracy and effectiveness, Choose and explain why¸ Rate, Rank, Defend, Choose, Grade, Order, Verify, Dispute, Criticize, Defend, Find the errors, Editorialize, Appraise, Judge, What fallacies, consistencies, inconsistencies appear, Which is more important, better, moral, appropriate, inappropriate, useful, clearer, suits the purpose, achieves the goal, logical, valid


Evaluation – Can students make JUDGEMENTS and support them?• Example of Evaluation Question• Disregarding the relative feasibility of the following

procedures, which of these lines of research is likely to provide us with the most valid and direct evidence as to revolutionary relations among different species? a) Analysis of the chemistry of stored food in female

gametes

b) Analysis of the form of the Krebs cycle

c) Observation of the form and arrangement of the endoplasmic reticulum

d) Comparison of details of the molecular structure of DNA

e) Determination of the total protein in the cell • Sources:

– http://www.nwlink.com/~donclark/hrd/bloom.html– http://www.fctel.uncc.edu/index.html


http://www.nwlink.com/~donclark/hrd/bloom.html

http://www.fctel.uncc.edu/index.html

Skills (Psychomotor)


Sample Practice Questions

1. Do you eat breakfast?

____ always

____ sometimes

_____never

2. How often do you eat breakfast during the week?

____ 1 – 2 days per week



____ everyday

____ never



3. What do you usually take to have a good breakfast? Give reason. (Please indicate number below).

FOODS REASONS

1. _____________________________________________

2. ____________________________________________

3. ____________________________________________

Reason: 1 – nutritious

2 – tastes good

3 – always available

4 – affordable

5 – others, specify ___________________



4. Why do you eat breakfast? (Please check all that apply to you.)

____ I feel better if I eat breakfast

____ I am hungry in the morning

____ my parents told me

____ I enjoy eating

____ I feel it is important to eat breakfast

____ others, specify ___________________



5. If you skip breakfast, how do you feel at mid-morning (9:30-10:00)? Check one that best describes how you feel.

_____ Fine

_____ Angry/bored

_____ Hungry/starving

_____ Headache/sick

_____ Nervous

_____ Easily irritated

_____ Weak

_____ Others, specify: ____________


Attitudes (Affective)


Sample Attitude Items• I cringe when I have to go to math class.

• I am uneasy about going to the board in a math class.

• I am afraid to ask questions in math class.

• I am always worried about being called on in math class.

• I understand math now, but I worry that it's going to get really difficult soon.

• I tend to zone out in math class.

• I fear math tests more than any other kind.

• I don't know how to study for math tests.

• It's clear to me in math class, but when I go home it's like I was never there.

• I'm afraid I won't be able to keep up with the rest of the class.


Put a check to the number which best corresponds to your answer:

1 – strongly agree 4 - disagree

2 – agree 5 – strongly disagree

3 – neutral

1 2 3 4 5

1. Adolescent who eats a wholesome nutritious breakfast do better in school.

2. I prefer to eat breakfast at home to have quality time with my family

3. Drinking milk is of much important as part of adolescent breakfast.

4. Skipping breakfast could lose out on an important nutritional contribution to adolescent’s total daily food intake.

5. Adolescents who take enough exercise will reduce the danger for excessive intake of calories.


6. I know that eating breakfast will improve my attendance and punctuality in class.

7. I believe that if I am able to consume a balanced meal during breakfast, I will have a “healthy day”.

8. If I eat breakfast, it is more likely that I will get the recommended dietary allowance for vitamin A, C, D, and B6, calcium, iron, zinc & folate.

9. If I have “instant noodle”, biscuit or coffee for breakfast, it will be enough to start my day.

10. Washing my hands before eating will keep me safe from food borne disease.

Put a check to the number which best corresponds to your answer:

1 – strongly agree 4 - disagree

2 – agree 5 – strongly disagree

3 – neutral


Table of SpecificationsContent area Recall /

RecognitionSkills Comprehension/ Application

Critical Thinking/ Problem Solving

Total Allocation of marks

Biochemistry 3 12 0 15

Cells/Tissues 4 13 3 20

Genetics Reproduction

2 10 3 15

Invertebrates 4 6 0 10

Vertebrates 5 11 4 20

Plant life 2 6 2 10

Ecological 0 7 3 10

Total 20 65 15 100Lizamarie Campoamor-Olegario

ITEM WRITINGStep 3.


Step 3. Item Writing

• 1. Keep the Test Short– Try out a longer test on a small number of

subjects, and then cut those sections that don't result in useful data.

• 2. Plan to Quantify Your Results.• 3. Good tests must be based on the

objectives of a course. • 4. Each test should meet four criteria:

RVDN– Reliability– Validity– Discrimination– Non-triviality


RVDN

• R: Does not generate random answers and consistently results in marks that reflect the skill level of each student.

• V: Measures what it intends to measure. • D: Shows clearly the differences in the

levels of achievement of the students to whom it is administered.

• N: Focuses on that which the students should know and not on that which is irrelevant to evaluation.


Step 3. Item Writing

5. Questions must be clearly written.

6. Least complex questions should appear first to encourage students' confidence in their knowledge.

7. Choose the test format. Commonly used are:a) essay

b) true-or-false tests

c) multiple choice tests

d) completion-and-short answer tests

e) matching tests

8. Reread all the items from the standpoint of the students


Essay Test

• Questions should clearly delimit the scope of the answer required

• Avoid words such as what, who, when, which trigger closed responses

• Avoid letting them to answer a choice of questions

• Give definitive task to student-compare, analyze, evaluate, etc.

• Use checklist point system to score with a model answer: write outline, determine how many points to assign to each part

• Score one question at a time-all at the same time.


True-or-false tests

• Avoid having more than one idea in a question.

• Avoid absolute terms: never, only, all, none, always

• Avoid indefinite terms: in most cases, great, sometimes, generally, some, few

• Avoid double negatives. • Use exact quantitative language • Don't lift items straight from the book. • Make more false than true (60/40).


Multiple choice tests

• In order to measure insight, avoid quoting from the textbook.

• Avoid “none of these” or “all of the above” • Stem should be in simple, understood

language; delete extraneous words. • Make all distracters plausible• Make all alternatives homogeneous (in subject,

content, form, length, explicitness, and grammatical structure)

• Don't use double negatives. • Present alternatives in logical or numerical

order.Lizamarie Campoamor-Olegario

Multiple choice tests

• Make each item independent of others on test. • Need more than 3 alternatives, 4 is best. • Ask for the best answer and use terms such as

“most” and “primary” if more than one answer is partially correct.

• The reading and linguistic difficulty of items should be low.

• Use care in the repetition of words or phrases between the stem and the correct answer.

• Avoid items that reveal the answer to another item.• Ordinarily, distracters should not overlap, subsume,

or be synonymous with one another.


Completion and short answer tests/ Identification• Limit the number of blanks in each question. • Generally, blanks are not inserted at the beginning

of the question. • Make sure there is only one unambiguous response

for the blank. • In numerical problems, indicate the type of units in

which the answer should appear. • State the item so that only one answer is correct• State the items with words that students understand• Avoid items in which the correct answer is a matter

of opinion.


Matching tests

• Organize questions together in logical groups.

• In your instructions, indicate whether or not some answers may be used more than once.

• Need 15 items or less. • Use homogenous material in each exercise. • Make all responses plausible. • Put all items on a single page. • Put response in some logical order

(chronological, alphabetical, etc.). Lizamarie Campoamor-Olegario

CONTENT VALIDATIONStep 4.


Step 4. Content Validation including Factor Analysis• Validity denotes the extent to which

an instrument is measuring what is supposed to measure.– Forms of Validity– Face Validity– Content Validity– Criterion-related Validity– Construct Validity


Face Validity

• Does the test appear valid? • Subjective judgment of a group

generally composed of non-experts• Includes good layout and format,

clarity of directions/ instructions, non-ambiguity of items, correct grammar and sentence structure(syntax), appropriate scales, appropriate number of items, clarity of printing


Exercises on Face Validity

Other than infancy, nutrition needs are higher in ________________ than at any other time in the life cycle because this is the period of rapid growth.a) Infancy

b)Adolescence

c) Adulthood

d)no response

e)do not know



__________ is the meal that should not be missed. Because it breaks the fast of not eating 8 to 12 hours.a)Breakfastb)Lunchc)Supperd)no responsee)do not know



The following food groups would consist of a good breakfast. a)coffee, biscuit, instant noodles

b)fried rice, fried dilis, tomato, banana, milk

c)peanut butter, pan de sal, coffee

d)no response

e)do not know



Eating breakfast has many benefits. It improves concentration, increase alertness, and helps one function more efficiently. a)true

b)false

c)no response

d)do not knowLizamarie Campoamor-Olegario


Adolescents who skip breakfast have no better academic performance and have negative effects on mood, alertness and contentment. a)true

b)false

c)no response

d)do not knowLizamarie Campoamor-Olegario


Adolescents need to consume at least ________ piece of fruit during breakfast..a)1 piece

b)2 pieces

c)3 pieces

d)No response

e)Do not know



Uriin ang mga sumusunod na pagkain kung ito ay isda o pagkaing dagat, uri ng prutas, o uri ng gulay. Ilagay lamang sa patlang ang letra ng tamang sagot at lagyan ng (/) ang hanay kung ang pagkaing ito ay Yamang Lupa o Yamang Tubig.



Upang maging malakas at masigla, dapat ay maging aktibo. Lagyan ng star (*) ang mga gawain na nakapagpapalakas ng katawan at (x) kung ito ay hindi masyadong nakapagpapalakas ng katawan.


Oo o Hindi. Ilagay sa puwang ang Oo kung ito ay tumutugon sa kakulangan sa pagkain o Hindi kung ito ay hindi makakatulong upang matugunan ang pangangailangan ng mag-anak sa pagkain

______1. Ang tatay ay nangingisda sa dagat.

______2. Ang mag-anak ay naglilinis ng bahay

______3. Ang tatay ay nagpuputol ng kahoy upang gamitin sa pagluluto.

______4. Ang mag-anak ay nagtatanim ng gulay sa paligid

______5. Ang bata ay tumatakbo ng matulin

______6. Ang pamilya ay nag-aalaga ng hayop

_____10. Ang mag-anak ay nagsisimba tuwing Linggo



Multiple Choice. Read the questions carefully and write the correct letter of your answer on the blank. (Basahin ang sumusunod na mga katanungan at ilagay sa blanko ang letra ng tamang sagot)

______ 1. Needed in the delivery of nutrients and removal of waste from the body (Ito ay kailangan sa pag-deliver ng sustansiya at pagtanggal ng mga dumi sa katawan) a) Rice (kanin)

b) Chicken (manok)

c) Fluids (tubig)

d) Fruits (prutas)



______2. To be more active, we should eat plenty of Go foods such as: (Upang maging aktibo, dapat tayo ay kumain ng “go foods” tulad ng): a)Bread, rice, cereals (tinapay, kanin at

cereals

b) fruits and vegetables (prutas, at gulay)

c) meat, fish, and chicken (karne, isda at manok)



______3. Grow Foods are important for growth and development, examples of grow foods are: (Ang “grow foods” ay importante para sa paglaki at pag-“develop”ng muscles) a)Bread, rice, cereals (tinapay, kanin at

cereals)

b) fruits and vegetables (prutas at gulay)

c) meat, fish, and chicken (karne, isda at manok)



______5. Breakfast is an important meal. Choose a good breakfast from the following. (Mahalaga na kumain ng agahan. Alin sa mga ito ang mainam na agahan?)a)Rice, dilis, banana and a glass of

milk

b)Bread and coffee

c)Noodles only Lizamarie Campoamor-Olegario


______9. Height and weight of a growing child should be taken every: (Ang taas at sukat ng katawan ng lumalaking bata ay dapat kunin kada):a)Month (buwan)

b)Year (taon)

c)Week (Linggo)


Matching Type. Match item A with Item B. Write your answers on the blank provided.

A B

_____ 1. Needed for proper hydration a) banana

_____ 2. Organ which produces insulin b) stomach

_____ 3. Where solid waste in the body is stored

c) digestive system

_____4. High potassium content d) water and other

fluids

_____ 5. Cause of diabetes e) lack of insulinLizamarie Campoamor-Olegario

_____ 6. The long tube between the mouth and stomach

f) mouth

_____ 7. The first part of the digestive system, where food enters the body

g) pancreas

_____ 8. Sac-like muscular organ that is attached to the esophagus

h) esophagus

_____ 9. Promotes health i) proper diet and physical activity

_____ 10. System that consists the organs of digestion

j) rectumLizamarie Campoamor-Olegario

II.Put a check (/) on the diseases of the digestive system if it is caused by the intake of dirty foods or (x) if not

______ 1. Flu ______ 6. Diabetes

______ 2. Diarrhea ______ 7. Sore throat

______ 3. Cough ______ 8. Dysentery

______ 4. Fever ______ 9. vomiting

______ 5. Stomach painLizamarie Campoamor-Olegario


What principle of healthy eating is indicated in the following sentences? Put your answer on the blank provided before the number. (Anong mga prinsipyo, na nakasalungguhit, sa pagkain ng wasto ang isinasaad ng mga pangungusap sa ibaba. Ilagay ang iyong sagot sa patlang. a)Balance

b)Variety

c) Moderation Lizamarie Campoamor-Olegario

1. Eating different kinds of food everyday is important to get all the essential nutrients needed by the body. (Ang pagkain ng iba’t ibang pagkain ay nakapagbibigay ng kumpletong sustansiya na kailangan ng katawan),

2. Eating the right amount of go, grow and glow foods (Ang pagkain ng tamang dami mula sa go grow and glow foods)

3. Put less sodium such as vetsin, toyo, patis in food to prevent high blood pressure. (Maglagay lamang ng kaunting dami ng vetsin, toyo,o patis sa pagkain upang maiwasan ang pagkakaroon ng mataas na presyon ng dugo)

4. A meal should consist of rice, protein source such as fish/meat, a piece or slice of fruit, at least half cup of vegetable and a glass of milk (Ang pagkain ay dapat na mayroong kanin, piraso ng karne, isang piraso ng prutas, kalahating tasa ng gulay at isang basong gatas.

5. Too much candies, cakes and chocolates may cause you to have tooth ache. (Ang pagkain ng sobrang matatamis tulad ng kendi, keyk at tsokolate ay nakapagdudulot ng sakit sa ngipin)


Identify the correct locomotor movement

_________ 1.A regular succession of leaps a. Run

_________ 2.To advance or move forward on foot with one foot always on the ground floor

b.Jump

_________ 3. To move fast with both feet off the ground

c.gallop

_________ 4. To slip or pass along over a surface d. Slide

_________ 5. to spring from the floor or ground f. LeapLizamarie Campoamor-Olegario

Multiple Choice. Write your answer on the blank provided before the number.

___________ 1. These are signs of stress?

a) healthy glowing skin

b) fatigue, dizziness and tension

c) mental alertness____________ 2. Gives the red color of the blood

a) Oxygen

b) Hemoglobin

c) Carbon dioxide____________ 5. Diabetes is caused by lack of:

a) Exercise

b) Insulin

c) Food Lizamarie Campoamor-Olegario

____________ 6. This carries nutrients and oxygen to the body

a)

b)

c)

Blood

Water

Fats

____________ 7. High blood pressure is caused by:

a) Excessive intake of fat

b) Excessive intake of salt

c) Excessive intake of sugar

___________ 8. High intake of fat increases the risk of:

a)

b)

c)

juvenile delinquency

heart disease

urinary tract infection

____________ 9. Over consumption of sodium leads to a high blood pressure. Foods high in sodium are:

a)

b)

c)

cakes

table salt, patis and bagoong

fruits


Content Validity

• The degree to which the sample of items, tasks, or questions on a test are representative of some defined universe or domain of content.

• It therefore, indicates whether or not the test items adequately represent that domain.

• Usually a table of specifications is built in order to ensure that the entire domain is represented by the items in a test.

• Expert judgment


Content Validity

• Is the instrument comprehensive? What is not there?

• Is it logical• How adequately does the sample of

items or questions represent the content to be assessed?

• Consult some experts who rate the items, eliminating or changing those that do not meet the specified content.

• Repeat until all raters agree on the questions and answers.


Criterion-related Validity

• The relationship between the scores obtained using the instrument and the scores obtained using one or more other instruments or measures


Two Types of Criterion-Related Validity• Concurrent Validity

• Predictive Ability


Concurrent Validity • Measure and criterion are assesses at

about the same time• Looks at information gathered from two

sources at the same time.• Example• New intelligence test and established test,

math ability and aptitude for engineering, student evaluation instrument and peer-evaluation instrument, teacher-made test and standardized test on the same subject area


Predictive Ability

• Used to look at validity over time (months/ years)

• Administer one assessment, allow time to lapse, then compare the earlier score to another measure of performance.

• Examples• The Scholastic Aptitude Test (SAT)

scores and GPA• SAT and Achievement Test• IQ test, GPA, and Achievement Test


Construct Validity

• Concerned with the extent to which an instrument may be said to measure a theoretical construct or latent trait

• Encompasses other elements of validity.

• Construct Validity Methods– Convergent validity– Divergent validity– Discriminant validity


Convergent Validity

• Correlating with other theoretical measure with which the developing instrument should correlate

• Example: • A new instrument to measure

professionalism in teaching and Teaching Commitment Scale

• (expect high correlation/ validity coefficient)


Divergent Validity

• Correlating with other theoretical measure with which the developing instrument should not correlate

• Example• A new instrument to measure

professionalism in teaching and Attitude Toward Movies Scale

• (expect low or negative correlation/ validity coefficient)


Discriminant Validity

• Examine the difference in test scores arising from groups known to differ on the construct

• Known-groups technique: comparison between known high and known low performers; difference must be significant; t-test


Construct Validation by Factor Analysis (Factorial Validity) • Tests that purports to measure the

same construct should load on the same factor, and that different factors should emerge for different constructs

• Items that load on the wrong factor are deleted or changed.


Factor Analysis – SPSS

• To run a factor analysis in SPSS, select Analyze (or Statistics) in the top menu bar.

• Move down to Data Reduction and select Factor Analysis.

• In the Factor Analysis dialog box, enter all the variables listed above in the "Variables" box.



• Click on the Descriptives button and check Coefficients, and Significance Levels.

• Click on the Extraction button and under Display check Unrotated Factor Matrix and Scree Plot. Leave as defaults the settings for Analyze Correlation Matrix and Extract Eigenvalues over 1.



• How many factors?• Perform factor analyses with various

values.• Choose the one that gives the most

appealing structure.• Click on the Rotation button and

select Varimax. Under Display, check Rotated Solution and Loading Plots.



• Click on the Scores button and check “Display factor score coefficient matrix".



• Click on the Options button and check "Coefficient Display Format, Sorted by Size".

• Click on OK to run the procedure.



• When you have a highly interpretable solution, which represents almost 90% of the data, name the factors.

• Look for patterns of similarity between items that load on a factor.

• better interpretation.


Example: Oblique Promax rotation of 4 factors of 24 mental ability variables (Gorsuch, 1983)

Factors 1 2 3 4General information .80 .10 -.01 -.06Paragraph comprehension .81 -.10 .02 .09Sentence completion .87 .04 .01 -.10Word classification .55 .12 .23 -.08Word meaning .87 -.11 -.01 .07Add .08 .86 -.30 .05Code .03 .52 -.09 .29Counting groups of dots -.16 .79 .14 -.09Straight & curved capitals -.01 .54 .41 -.16Woody-McCall mixed .24 .43 .00 .18Visual perception -.08 .03 .77 -.04Cubes -.07 -.02 .59 -.08Paper form board -.02 -.19 .68 -.02Flags .07 -.06 .66 -.12Deduction .25 -.11 .40 .20Numerical puzzles -.03 .35 .37 .06Problem reasoning .24 -.07 .36 .21Series completion .21 .05 .49 .06Word recognition .09 -.08 -.13 .66Number recognition -.04 -.09 -.02 .64Figure recognition -.16 -.13 .43 .47Object-number .00 .09 -.13 .69Number-figure -.22 .23 .25 .42Figure-word .00 .05 .15 .37



• High values (above about .4 in absolute value)

• Variables within each block all seem to measure the same general kind of mental ability

• Note: Criterion and Construct Validation can be done after Pilot-testing


PILOT-TESTING TO EQUIVALENT GROUP

Step 5.


Step 5. Pilot-testing to equivalent group • If achievement test, to higher level• Number of items times three


ITEM ANALYSIS, RELIABILITY

Step 6.


Step 6. Item-analysis, reliability

• To improve the effectiveness of test items and the validity of test scores

• Select and rewrite items on the basis of item performance data

• Item analysis provides information about:– The difficulty of the item: How easy or hard

the item was

– The discriminating power of the item: How well it discriminated between high and low scorers in the test

– The effectiveness of the alternative: Whether all the alternatives served their purpose


Steps of an Item Analysis

• S1. Order the N papers by score, placing the one with the highest score on top and continuing sequentially until on with the lowest score is on the bottom. – Note: Where two or more

examinees made the same score, order them alphabetically by surname.



• S2. Multiply N, the total number of tests, by 0.27 and round off the result to the nearest whole number, the number is represented by n. If N is 30, n would be 8 (8.1 rounded)

• S3. Count off the n best papers from the top of the stack. This is the “high” group.

• S4. Count off the n poorest papers from the bottom of the stack. This is the “low” group.



• S5. Determine the proportion in the high group (pH) answering a particular item correctly by dividing the number of correct answers for the high group by n, that is: Number of correct responses to the item

pH = ----------------------------------------------------------

n

• Repeat the procedure for the low group to obtain pL for each item



• S6. To obtain an estimated item difficulty index, p (that is, the proportion of the total group that answered the item correctly), add pH to pL and divide the resulting sum by 2: pH + pL

p = ---------------

2


• S7. To obtain a measure of item discrimination (i.e., how well this item distinguishes between the students who understand the content universe of the test and those who do not), subtract pL from pH:

• D = pH-pL

Item Number

High or Low

Number of Correct Responses for Groups (n=8)

Proportion of Correct Responses

Item Discrimination

D = pH-pL

Item Difficulty

pH+pL

p = --------

2

1 H 7L 3

2 H 8L 6

3 H 4L 4

4 H 2L 6

n = (0.27) Nn=0.27 (30) = 8.1 (rounded off to 8) students in the high group and 8 in the low groupLizamarie Campoamor-Olegario

Item Number

Group Number of Correct Responses for Groups (n=8)

Proportion of Correct Responses

Item Discrimination

D = pH-pL

Item Difficulty

pH+pL

p = --------

21 H 7 pH=7/8=0.8

80.88-0.38 = 0.50

0.88+0.38

2

=0.63

L 3 pL=3/8=0.382 H 8 1.00 0.25 0.88

L 6 0.753 H 4 0.50 0 0.50

L 4 0.504 H 2 0.25 -0.50 0.50

L 6 0.75

Answer



• For discrimination, the following table may be used:– 0.40 and above = very discriminating

item– 0.30-0.39 = discriminating item– 0.20-0.29 = reasonably discriminating

item– 0.10-0.19 = marginally discriminating

item, usually subject to improvement– Below 0.10 = unproductive item for test

reliability purposesLizamarie Campoamor-Olegario

Difficulty


Interpreting Item Analysis

• Good item• Potential miskey• Ambiguous items• Equal distribution to all alternatives• Alternative/s not working• Distracter too attractive• Question not discriminating• Negative discrimination• Too easy• Omit


Good item

• * keyed correct answer• O omit or no answer

Group A B C* D O

High 0 0 10 0

Low 4 3 7 2


Potential MiskeyGroup A* B C D O

High 1 1 10 3

Low 4 3 5 3


Ambiguous ItemsGroup A B C* D O

High 6 1 7 1

Low 5 2 5 3


Equally to all alternatives

Group A B C* D O

High 4 3 5 3

Low 3 4 4 4

Students are guessing


Alternatives aren’t working

• Each distracter has to attract at least 5% of the students

• If two alternatives don't draw any students --> might consider redoing as true/false

Group A B* C D O

High 1 14 0 0

Low 5 7 2 0


Distracter too attractive

• Not one distracter should pull more than about half of students

Group A B C D* O

High 7 1 1 5

Low 10 2 1 2


Question not discriminating

Group A* B C D O

High 7 3 2 3

Low 7 2 1 5


Negative Discrimination

Group A* B C D O

High 7 3 2 3

Low 10 3 1 1

usually means an ambiguous question


Too Easy Group A B* C D O

High 0 14 0 1

Low 1 13 1 1

If you dropping too many questions, you may lose content validity (specs)


Omit

• Possibilities– If near end of the test, they didn't find it because

it was on the next page (format problem)– Test is too long– Way too difficult

Group A B* C D O

High 0 14 0 1

Low 1 13 1 1


Reliability Analysis

• Reliability is consistency of test scores• The difference of test scores is attributed to

“true differences” • Not to

– testing environment– instruction– time limits– rapport– fatigue– practice– examinees mood– and other similar factors


How can these be avoided

• Careful planning of the test • Good sampling of the items


Reliability Coefficient

• Used in determining the consistency of tests

• The larger the number is, the more consistent the assessment

• .90 or higher means high reliability• .80 to .89 good reliability• .60 to .79 means low to moderate

reliability • .40 to .59 means doubtful reliability


Internal Consistency

• Cronbach’s Alpha• a measure of the “here-and-now, on

the spot” reliability• Limitations of internal consistency

include– 1) there needs to be sufficient

number of items (around five items per trait or skill)

– 2) it is not appropriate for speeded tests.


Other Things to Look at

• Number of items– Cronbach's alpha increases as the

number of items in the scale increases.

• Alpha if deleted– The researcher may wish to drop

items where the alpha if deleted is higher than the overall alpha as another way to improve the alpha level.


Other Things to Look at

• The item-total correlation– A low item-total correlation means

the item is little correlated with the overall scale (ex., < .3 for large samples or not significant for small samples) and the researcher should consider dropping it.

– A negative correlation indicates the need to recode the item in the opposite direction.


Remember

• The reliability analysis should be re-run if an item is dropped or recoded.


In SPSS, Cronbach's alpha is found under Analyze, Scale, Reliability Analysis


Then in the Statistics button, check Scale to get alpha. You can also check Scale if deleted


R E L I A B I L I T Y A N A L Y S I S - S C A L E (A L P H A)

Item-total Statistics

Scale Scale Corrected

Mean Variance Item- Alpha if Item if Item Total if Item

Deleted Deleted Correlation Deleted

ITEM 1 60.6978 115.0779 .3427 .9033

ITEM 2 60.6400 117.0439 .0765 .9047

ITEM 3 60.6800 116.0936 .2072 .9041

ITEM 4 60.7111 115.0724 .3248 .9034

ITEM 5 60.7422 114.7547 .3335 .9033

ITEM 6 60.7956 116.4223 .0989 .9051

ITEM 7 60.6844 115.1098 .3602 .9032

ITEM 8 60.6711 115.3646 .3441 .9034

ITEM 9 60.6756 116.4702 .1497 .9044

ITEM 10 60.7911 114.2999 .3485 .9031

N of Cases = 225.0 N of Items = 80

Alpha = .9045


FINAL ADMINISTRATIONStep 7.


Step 7. Final Administration

• Usually, the run is five years


Thank you for listening!


test construction and evaluation lizamarie campoamor-olegario

Documents

definition of construct

factor analysis step

reliability analysis

example of knowledge

cognitive knowledge

equivalent group step

type of test

test construction