Development of Norms this is presentation.pptx

Presentation related to norms

1. Development of Norms and Grades Dr. Pratibha Sagar

2. Basic Statistical Concepts: Measures of Central tendencies: Mean Average Median Middle value Mode Most repeated score Measures of Variability: Standard deviation Quartile deviation Range Z Scores

3. Why interpreting raw score? Student A Math = 50 English = 40 Science = 70 Student B Math = 60 English = 45 Science = 40

4. Why interpreting raw score? A raw score on any psychological test is meaningless. Raw score Percentage The difficulty level of the items making up each test will determine the meaning of the score.

5. How to interpret scores? Relative Score (Norm-referenced) Absolute Score (Criterion-referenced)

6. Norm s Stan Standard/Average performance . Methodology – to understand psy tests and proper interpretation of scores. Norms, Reliability , Validity, Item Analysis and Test Design. Raw Scores - 35 in English. 20 in math ???? Expressed in different units –kg, hour , no .of correct/incorrect responses, no. of trails, So we cant just directly compare it. It can only interpreted in clearly defined and uniform frame of references .

7. Norm s Stan Definition: “ Norms may be defined as the average performance on a particular test made by a standardization sample .” Standardization sample-true representation population, cross cultural, Empirically established . To discover where S/he falls in that situation - we convert Raw scores into Derived Score-

8. Norm s Stan Definition: Derived Score- Difficulty level – all score low 1,5,10 /1000. we can compare D score objectives /purpose/ goals of Derived Score . To indicate the individual’s relative standing in the normative sample and thus permit an evaluation of her/his performance in reference to other persons. To provide comparable measures that permits a direct comparison of the individuals performance on different tests.

9. Norms Standard/Average performance . Methodology – to understand psy tests and proper interpretation of scores. Norms, Reliability , Validity, Item Analysis and Test Design. Raw Scores - 35 in English. 20 in math ???? Expressed in different units –kg, hour , no .of correct/incorrect responses, no. of trails, So we cant just directly compare it. It can only interpreted in clearly defined and uniform frame of references .

10. Norms Steps in developing Norms: Defining the target population Normative group -based on intention of test . Selecting the sample from the target population True representative sample. Cross sectional Large sample Random sampling Standardizing the conditions Test administration must be standard, valid

11. Developmental Norms How far along the normal developmental path the individual has progressed They are psychometrical crude and do not lend themselves well to precise statistical treatment.

12. Developmental Norms Mental Age Grade Equivalent Ordinal Scale

13. Mental Age: Age-equivalent items Items were grouped into year levels. Scores based on the highest year level that child attained. In actual practice, the individual performance showed certain amount of scatter. Therefore, it was customary to compute basal age and give partial credits for all test passed at higher year levels.

14. Mental Age: Age norms Standardized samples of each age constitute the age norm on the test. All raw scores on such a test can be transformed in a similar manner to the age norms.

15. Mental Age: Limitations The mental age unit tend to shrink with advancing years. Intellectual development progress and maturity

16. G r ad e -equi v ale n t Like Mental Age Computing the mean raw score obtained by children in each grade and then equating raw score to grade mean

17. Grade-equivalent: Limitations Content of instruction varies from grade to grade  Unequal distance Grade norms tend to be incorrectly regarded as performance standards Grade norms are subject to misinterpretation

18. Four levels or scales of measurement Nominal scales – involve classification or categorization based on one or more distinguishing characteristics, Ex - “men,” “1,” “B,” or “women,” “2,” or “A.” Ordinal scales- permit classification and rank ordering on some characteristic Ex- merit list of SP College .

19. Four levels or scales of measurement Interval scales - contain equal intervals between numbers. But like ordinal scales, interval scales contain no absolute zero point Ex - IQs of 80 and 100 Ratio Scales In addition to all the properties of nominal, ordinal, and interval measurement, It scale has a true zero point.

20. Within-group Norms The individual’s performance is evaluated in terms of the performance of the most nearly comparable standardized group. Clearly quantitative meaning defined

21. Within-group Norms Percentile: percentage of persons in the standardization sample who fall below a given raw score Standard Score (z-score): transformation from raw score to M = 0, SD = 1 Derived Standard Score: transformation from raw score Deviation IQ: M = 100, SD = 15 or 16 (depend on tests) T-score: M = 50, SD = 10 CEEB score: M = 500, SD = 100 Stanine: M = 5, SD = 2

22. Within-group Norms Percentiles: Most common and popular Percentile -% of persons (standard sample) fall below a given point. Percentile and Percentile Rank are two different concepts. ex – if the 30% of the person obtain fewer than 18 problems correct on math then raw score of 18 corresponds to 30 th percentile (P 30 ) i.e. percentile rank is 30 and percentile score is 18 Lower the percentile the poorer the persons standing.

23. Within-group Norms PR 50 –median .PR 25 and PR 75 are called 1 st n 3 rd quartile points .. Different from percentage ( %) – raw score where percentile is derived score. PR 0 & PR 100? a raw score lower/more than obtained in in the standard sample.

24. Within-group Norms  Standard score and derived standard score  Linear Transformation  Normalized Transformation Raw score S tanda r d score Raw score Cumulative percent Standard Score

25. Within-group Norms Why transform to normal distribution? Most raw score distributions approximate the normal curve more closely than they do any other type of curve. Normal curve has many useful mathematical properties. Normalized transformation should be carried out only when the sample is large and representative. Deviation from normality results from defects in the test