data_analysis

Slide 1: 

DATA ANALYSIS Presentation Teaching & Learning Commons Presentation http://talc.ukzn.ac.za

Slide 2: 

DATA ANALYSIS Quantitative Analysis Qualitative Analysis Establishing Validity and Reliability of the findings Slide # 2 http://talc.ukzn.ac.za

Slide 3: 

Slide # 3 http://talc.ukzn.ac.za

Slide 4: 

Slide # 4 QUANTITATIVE ANALYSIS Summarisation of Data Coding Percentages Mean Standard Deviation Histogram and other Graphs Interpretation of Scores obtained by Students in exams Establishing Validity and Reliability of the findings http://talc.ukzn.ac.za

Slide 5: 

Slide # 5 SUMMARISATION OF DATA Avoid Intermediate Recording Systems More times data are transferred, the greater is the risk of introducing errors If possible, design the data collection instrument (Questionnaire) so that the supplied responses can be processed by computers directly (i.e. without incorporating a coding stage) http://talc.ukzn.ac.za

Slide 6: 

Slide # 6 DATA CODING SHEET DATA PERTAINING TO __________________ http://talc.ukzn.ac.za

Slide 7: 

Slide # 7 SEX ID H.E.Q. T.E. http://talc.ukzn.ac.za

Slide 8: 

Slide # 8 CODING Case Identification number Column Row Fixed-Format Coding The same item of information appears in the same column for all cases Digits and Width Order / Sequence Case I.D. Number Bio-Graphic Data Research / Evaluation Data Spacing Blank Space Decimal Point Dichotomous Data Zero Code http://talc.ukzn.ac.za

Slide 9: 

Slide # 9 FREQUENCY DISTRIBUTION Determine the interval size Establish the upper and lower limits of the interval Establish all other intervals Record the frequency in each interval http://talc.ukzn.ac.za

Slide 10: 

Slide # 10 HISTOGRAM A series of columns (BARS), each having as its base one class interval and as its height the frequency in that class Frequency Polygon Frequency Curve Which Graph is best? http://talc.ukzn.ac.za

Slide 11: 

Slide # 11 MARKS (CLASS INTERVALS) FREQUENCIES HISTOGRAM http://talc.ukzn.ac.za

Slide 12: 

Slide # 12 MARKS (CLASS INTERVALS) FREQUENCIES FREQUENCY POLYGON http://talc.ukzn.ac.za

Slide 13: 

Slide # 13 YEAR PERCENTAGE OF STUDENTS PASSING IN MATHS TIME SERIES ANALYSIS http://talc.ukzn.ac.za

Slide 14: 

Slide # 14 ANALYSIS OF THE RESPONSES TO OPEN ENDED QUESTIONS Take one question at a time and study all the responses to that question Perform “Content Analysis” Look for Recurring Regularities Figure out what things fit together (convergence) Identify patterns and label the various types of information Sort the information into categories (i) Contd. http://talc.ukzn.ac.za

Slide 15: 

Slide # 15 ANALYSIS OF THE RESPONSES TO OPEN ENDED QUESTIONS (ii) Judge the categories by using two criteria: Internal Homogeneity Extent to which the information in a category hold together Extent to which differences among categories are bold and clear External Heterogeneity http://talc.ukzn.ac.za

Slide 16: 

Slide # 16 Slide # 16 PROBLEM OF DIVERGENCE Flesh out the categories by using the following: Building on items of information already known http://talc.ukzn.ac.za

Slide 17: 

Slide # 17 ESTABLISHING VALIDITY AND RELIABILITY OF THE FINDINGS This can be done by data “triangulation” i.e. by comparing the information collected from: Different Sources the same source through different methods of data collection the same source through different investigators http://talc.ukzn.ac.za

Slide 18: 

Slide # 18 Information collected from different sources Information collected from the same source through different methods Comparison of Information collected from the same source through different persons (Interviewers / Observers) TRIANGULATION http://talc.ukzn.ac.za

Slide 19: 

Slide # 19 Observational data will provide support for data obtained form other sources such as responses to questionnaires or interviews. For example, this may be necessary when the validity of teacher’s responses to questionnaires may be questioned on the basis of the “social desirability effect”. Further, 10 or 15 percent of the sample responding to the questionnaire may be interviewed in order to cross check the original responses. Contd. http://talc.ukzn.ac.za

Slide 20: 

Slide # 20 For estimating reliability, alternate forms of some critical questions can be asked in different parts of the same questionnaire. If the responses to the alternate forms of the same questions are similar in a majority of the cases then we can conclude that the information obtained is reliable. http://talc.ukzn.ac.za

Slide 21: 

15 The weightage assigned for Internal Assessment should be increased STRONGLY AGREE AGREE DISAGREE STRONGLY DISAGREE Slide # 21 ITEM No.: http://talc.ukzn.ac.za

Slide 22: 

Slide # 22 ITEM No.: 56 The weightage assigned for External Assessment should be increased STRONGLY AGREE AGREE DISAGREE STRONGLY DISAGREE http://talc.ukzn.ac.za

Slide 23: 

DATA ANALYSIS Dealing with actual facts uninfluenced by personal feelings / opinions Making impartial judgements Ability to free oneself from personal prejudice Slide # 23 OBJECTIVE SUBJECTIVE Objectivity Judgements based on personal impression Analysis of data is done by applying personal feelings / opinions (Not by using fixed external criteria) Subjectivity Vs http://talc.ukzn.ac.za

Slide 24: 

Slide # 24 hansrajhr@ukzn.ac.za http://talc.ukzn.ac.za Thank You