2-6-Data-Measurement research for nursing students

ppt about data measurement in research

1. Data Measurement

2. Advantages of Measurement Remove subjectivity and guesswork since measurement is based on rules, hence, finding can be verified. Reasonably help obtain precise information . Communicate numbers which is exact than words.

3. Errors in Measurement Measurement error is the difference between the actual amount of the attributes (true score) and the amount of the attribute that was measured (observed score).

4. Types of Measurement Errors

5. Random Error Expected and is affected by a host of influences that may be present in an experiment. due to human factors such as confusion, bias, and variation in procedure as well as data processing. does not affect the average scores of data set but affects the amount of variation that exists around the average scores.

6. Systematic Error seriously affects the results of a research study appears to be accurate, while it remains to be consistently biased disastrous

7. Sources of Measurement Errors

8. Environmental contaminants: temperature , lightning, time, friendliness of researchers. Variation in personal factors: fatigue, hunger, anxiety or mood. Quality of Responses: accurate measures, social desirability, acquiescence and extreme responses which can cause potential problems in self-report measures.

9. Variation in Data Collections: different ways of collecting data, instruction, coding categories, and approaches. Clarity of the Instrument

10. Sampling of ‘Questionnaire’ Items depends on what items are included as discussed and learned. Format of the Instrument Technical Aspect Open- VS closed-ended oral VS written

11. Strategies to Minimize Measurement Errors

12. Measurability or Reliability Reliability of instrument is the consistency with which an instrument measures the attribute. Reliability also measures accuracy of an instrument which shows true score and minimize the error component of the obtained scores. Reliability affects the precision of a measure.

13. a. Stability of Measurement Same samples on separate occasion. Test-retest reliability procedures. Compare the scores by computing the re liability coefficient which measures the magnitude of the test’s reliability. Proper calibration of physical instruments and keeping instruments in top working conditions and well calibrated with ensure stability of measurement.

14. b. Internal Consistency Shows that all indicators or subparts measures the same characteristics or attributes of the variables. Split half technique and Cronbach's alpha or coefficient alpha.

15. c. Equivalence High consistency or agreement and congruence in the observation ratings of the different observers or raters of certain phenomenon. Reliability estimates of 98 indicates high level correlation or agreement between observers.

16. 2. The Split-Half Technique Determine homogeneity of items. Items are split in half end core relational procedure performed between the two halves. Odd or even items are scored Reliability between the two will be high.

17. 3. The Cronbach's Alpha or Coefficient Gives an estimate of the split half correlation.

18. 4. Kuder-richardson (KR-20) Coefficient “Yes or no” or “true or false” response. Requires consistency of responses to all the items of a single form of a test that is administered at one time.

19. 5. Validity Degree by which the instrument measures what it intends to measure. Validity should prove that the instrument will consistently measure the right variables to be investigated. Face validity Content validity Face validity Criterion related validity Construct validity

20. Content validity- concerned with the adequacy of the content area being measured. Face validity- c oncept or variables desire by the researcher in the study. Criterion-Related Validity - some external criterion. the measurement is valid if it strongly corresponds with scores on the criterion and can predict attitude, behaviors, experiences among others.

21. Two Types of Criterion Related Validity predictive validity- ability to differentiate between the respondents’ performance or behavior on some future criterion outcome. Concurrent validity- ability to measure the differences in respondents present status on some criterion.

22. d. Construct Validity Reducing concept abstraction into a more concrete and suitable criterion related validation. Hypothesis testing approach is then used to validate the instrument. Gather data and test the bar being scores of the sample, make interferences based on the findings and determine the rational whether the construction of the instrument is adequate to explain the findings.

23. Sensitivity and specificity Sensitivity is the ability of the instrument to correctly screen or identify the variables to be manipulated and measured to diagnose its condition . Specificity is the ability of the instrument to correctly identify non-cases or extraneous variables and screen out those condition not necessary for manipulation.

24. Sensitivity and specificity Sensitivity is the ability of the instrument to correctly screen or identify the variables to be manipulated and measured to diagnose its condition . Specificity is the ability of the instrument to correctly identify non-cases or extraneous variables and screen out those condition not necessary for manipulation.

25. Assessing Qualitative Data

26. Assessing Qualitative Data “True state” of human experience. necessary to influence acceptance of authorities. Lincoln and Guba (1985) 5 Criteria

27. Credibility Confidence in the truthfulness of data and interpretations. There are various techniques that can improve in document credibility of qualitative data , and this is done by believability is enhanced .

28. Steps to Demonstrate Credibility Prolonged engagement - group and spend time to collect in that data by engaging their views and understanding their feelings. Persistent observations- are relevant to the phenomenon being studied to the point of saturation. Triangulation- multiple references to draw. Triangulation is to prevent the intrinsic bias that comes from the single method, single observation and a single theory studies. A complete and contextualize status of a phenomenon.

29. d. Peer debriefing and member checks- peers to objectively review and explore various aspects of the phenomenon. e. Search for disconfirming evidence – search for data could challenge emerging concept or descriptive theory. Proposed sampling and facilitated by prolonged engagement and debriefing.

30. Dependability Dependability is concerned with the stability of qualitative data for a long period of time, stepwise replication, conduct inquiry separately and then compare their data and conclusions.

31. Confirmability Objectivity and neutrality of data determine accuracy, relevance and meaning of data. Findings must reflect the participants’ opinion, feelings and attitude regarding the inquiry and not the opinions, bias or perspective.

32. Transferability Research findings which can be transferred or applied to other settings. Provide a thorough and sufficient descriptive data of the research setting. Process observed during the inquiry.

33. Strategies to Enhance Qualitative Data Analysis Intensive listening during an interview.

34. THANK YOU! Any questions?