Data Collection and Analysis community medicine.pptx

Data collection and analysis

1. Data Collection and Analysis Prepared by Dr. Hadi AL- Hamzi MBBS,MSC Community Medicine Department Faculty of Medicine and Health Sciences Amran University

2. Introduction Data collection is perhaps the most important component of Monitoring and Evaluation Without good data we can not tell if or not the project is beneficial as expected Mechanisms must be put in place to collect adequate, timely and relevant data on a continuous basis

3. Qualities of good data Adequate Timely Relevant Consistent Reliable

4. Data Collection methods Questionnaire Observation Laboratory processes Desk review (books, records, other documents)

5. Questionnaire Quality and adequacy of data collected depends on the nature of questionnaire used The questionnaire must address all the variables under study Length and complication (language) of the questionnaire can affect data collection

6. Types of Questionnaires Questionnaires may be : Open-ended Closed-ended

7. Open-ended Questionnaires Allow respondent to answer the question in his or her own words Example: What benefits have you drawn from the project ? Open-ended questions allow the interviewer to ask or probe the respondent further if he/she feels that clarification of a point, or additional information, is needed. Good for exploratory research. Disadvantage are: the difficulties of analyzing the data and in categorizing and summarizing answers because of the unique responses. Also, there is the possibility of interviewer bias and bias caused by the different education levels of the respondents

8. Closed-ended Questionnaires Respondent answers the question in in very specific pre-determined choice of words Example: Would you study participate in the project if you were to make the choice a 2nd time? (tick yes or no) They are easy to statistically analyze

9. Types of Questionnaire Administration Self administered Personally administered Group administered

10. Key questions when designing a questionnaire What should the respondents be asked? How should each question be phrased (formulated)? In what sequence should the questions be arranged? What is the best questionnaire layout for the research problem in question? What communication medium should be utilized (personal interview, telephone interview etc.) Should the questionnaire be pretested? How should the questionnaire be pretested? Does the questionnaire require a revision?

11. Relevance and Accuracy Questionnaires must collect data that is both accurate and relevant Relevance means: 1. No unnecessary information is obtained from the questionnaire 2. All the information that is needed for the purpose of the evaluation is collected 3. No important information is omitted Accuracy means that the questions are worded in a manner which ensures the collection of correct information from respondents

12. Guidelines for phrasing questionnaires Avoidance of Complexity / Use Simple Language Avoidance of leading and Loaded questions Counter biasing Statements (leading respondent with an introductory remark before the question Avoidance of Ambiguity غموض Avoidance of Double-Barreled Questions – Questions which address two or more issues simultaneously Avoidance of Assumptions Avoidance of Burdensome (and Memory Taxing) Questions

13. Sequencing Questions The order in which questions are put in a questionnaire may significantly affect the response rate Usually, interviewers prefer to ask general questions from respondents before moving on to specific questions (funnel technique)

14. Key Factors to consider in questionnaire layout Don‘t overcrowd the questionnaires Keep questionnaires as brief as possible Ensure that the title and subtitles of the questionnaire and questionnaire sections are carefully phrased and captures the respondents attention Use appealing language

15. Pretesting Questionnaires Pretesting is a very useful method for determining whether respondents have any difficulty understanding the questionnaire and whether the questions are ambiguous or can lead possibly to biased answers Pretesting ensures that costly errors in questionnaires which are given to a large number of respondents is avoided The respondents involved in a pretest should be similar in essence to the target respondents of the research

16. Pretesting provide answers to important questions for the, such as: Can the questionnaire format be followed by the interviewers? Is the questionnaires ambiguous? Does the questionnaire flow naturally and conversationally? Can respondents answer the questions easily? Which alternative forms of questions work best?

17. Data Analysis

18. Methods of Data Analysis Quantitative analysis Qualitative analysis Both methods can be applied manually or by computer software (e.g. SPSS or MS Excel)

19. Quantitative Analysis Analyses done through use of statistical processes that may include derivation of percentages, ratios, correlations and trends among others as may be required Statistics give a more accurate picture of the situation at hand Statistics must be selected carefully- they should be able to give expected results

20. Qualitative Analysis Not all data can be analyzed quantitatively Some data is too qualitative and so has to be analyzed by applying perceptional meaning Thematic analysis is one of the most popular methods of qualitative analyses In this method data is classified into themes based on its relatedness

21. Choice of Method Depends on: Nature of data to be analyzed Nature of expected findings Expected use of findings Audience of the findings A mixed method is preferred in most cases

22. End