Third Group Training Course in Application of Information and

1

• Third Group Training Course in Application of
  Information and
• Communications Technology to Production and
• Dissemination of Official Statistics
• (06th May07 12 Jul07)
• Compare Various Types of Interviewing Data
  Capturing Methods
• By
• Group - II
• Ms. Rashmi Sharma, India
• Mst. Maksuda Shilpi, Bangladesh
• Ms. Aminath Shirmeen, Maldives
• Mr. Amadou Tidiane Diallo, Senegal

2
Presentation Overview

• Introduction - Amadou
• Project Activity Plan - Shiru
• Questionnaire - Shiru
• Software Used - Rashmi
• Software Development - Rashmi
• Data Collection - Shilpi
• Tabulation - Shilpi
• Data Analysis Conclusions - Amadou
• References - Amadou

3
Introduction

• Target group is TIC participants
• Sample concerned with 54 individuals
• Sample distribution21 males and 33 females,
  drawn from all 5 regions
• Response rate is 96

4
Project Activity Plan and the Questionnaire
Shiru
5
Activity Plan

• Activity plan
• Meetings
• Choosing the Topic
• Design questionnaire
• Develop Data entry programme
• Data Collection and Data Entry
• Data cleaning and Analysis
• Finalize tables
• Dissemination in different format

6
Meetings

• We had total 7 meetings.
• We held our meetings in TIC lobby and one meeting
  during normal project class at SIAP.
• We changed our survey topic 3 times.
• 5th meeting We finalized survey topic.
• Meetings

7
Data Collection

• We used 4 methods of data collection
• PAPI
• CAPI
• CATI
• Electronic method using email

8
Factors considered during questionnaire design

• Flow of questions
• Multiple answers
• Skip questions

9
Questionnaire

• Design Questionnaire in excel format.
• Total no. of Questions 15
• No. of Multiple questions 3
• No. of Skip questions 2
• Questionnaire
• PAPI and CATI (questionnaire1)
• Electronic (questionnaire2)

10

Software Development
Rashmi
11
Software Used

• CSPro
• STATA SPSS
• Microsoft Access
• Microsoft Excel
• Microsoft Word
• Microsoft FrontPage
• Xerver

12
Data Entry Application

• Data entry app s/w developed in
• CSPro, Access, Excel
• Data file specification
• Number of Variables 37
• - Numeric variables 35
• Alphanumeric variables 2
• Number of record type 1

13
Created Date Entry Application

• Created Dictionary All items in CSPro dictionary
• Created Identification Identification, Id, 2
• Identify Records and variable, sub items
  (Person)
• Identify variable label Exact Question Text
• Identify variable name Qltquestion numbergt
• Size of variable (according to question)
• Type of variable (according to question)
• Identify Values (according to question)
• Fix number of occurrences (according default)
• Click here

14
Data Dictionary
15
Value Set
16
CSPro Data Capture Screen
17
Logic
18
CAPI Question
pop-up responses box
19
MS Excel

• Macro Recording
• Data save in ASCII form
• Align the data according to CSPro Data Dictionary
  format

20
MS ACCESS
21
Data Integration
CSPro PAPI
CSPro DATA File
ACCESS CAPI
Concatenation (CSPro)
CSPro CAPI
EXCEL EMAIL
Output Data
CSPro CATI
Input Data
22
Data Integration

• Integration of the different data files format
• The integrated file is handed over to all group
  members

23
Data Collection and Tabulation

• Mst. Maksuda Shilpi

24
Methods and Name of Collector

1. PAPI Methods by Shilpi
2. CAPI Methods by Rashmi
3. CATI Methods by Amadou
4. Electronic Methods by Shiru

25
Methods and Number of Cases

1. 10 Cases by PAPI Methods
2. 8 Cases by Electronic Methods
3. 19 Cases by CAPI Methods
4. 17 Cases by CATI Methods
5. 2 Non-responses in Electronic Methods

26
Data Integration

• Using CSPro data entry application all these
  files are concatenated.

27
Tabulation

• We made tabulation by CSPro Tabulation
  Application.
• No. of Table9
• No. of Frequency Table5
• No. of Cross Table4
• Demonstration of tabulation application

28
Table 1 Percentage Distribution of Method Used
Method Used Percent
PAPI 18
CAPI 35
CATI 32
Electronic 15
Total 100
29
Table 2 Percentage Distribution of Time to fill
Questionnaire
Time to fill questionnaire Percent
0 min- 5 min 59
6 min- 11 min 37
12 min- 17 min 2
18 min- 23 min -
gt23 min 2
Total 100
30

• Amadou

31
Data Analysis Conclusions

• Analysis is done under STATA, SPSS Excel
• Interpretation of Results
• Survey findings

32

• Tables converted from CSPro into SPSS and Stata
• Converted data copied and pasted into Excel and
  data ploted

33

• 53.13 of respondents completed the
  questionnaires through CAPI within 5 minutes
• 21.88 completed the questionnaire by PAPI
  within the same time interval.

34

• 15.63 (third of CATI proportion) filled in the
  questionnaires within the first time interval
• The proportion goes further down with CATI (9.38
  ).

35

• 15.63 (third of CATI proportion) filled in the
  questionnaire by Electronic within the first time
  interval
• The proportion goes further down with CATI (9.38
  ).

36

• 15.63 (third of CATI proportion) filled in the
  questionnaire by Electronic within the first time
  interval
• The proportion goes further down with CATI (9.38
  ).

37

• 8 respondents in 10, whose first language is not
  English, completed the questionnaire between 5
  and 11 minutes
• 2 in 10, whose first language is English,
  completed the questionnaire within the same
  interval

38

• 63.6 of female respondents completed the
  questionnaire within 5 minutes
• 52.4 of male respondents completed the
  questionnaire within the same time interval

39
Descriptive statistics under Stata
variable Obs Mean Std. Dev. Min Max
Method used 54 2.42 0.96 1 4
Region 54 2.70 1.10 1 5
Sex 54 1.61 0.49 1 2
Age group 54 1.89 0.69 1 4
Education 54 2.67 0.80 1 5
First language 54 1.85 0.35 1 2
Fill a questionnaire before 54 1.17 0.38 1 2
Time used 54 1.48 0.72 1 5

• Method used
• Mean 2.42 Standard deviation0.96 Minimum
  value1 and Maximum value4

40
Regression analysis under STATA
Source SS df MS Number of obs 54
        F( 7, 46) 1.66
Model 5.53425486 7 0.79060784 Prob gt F 0.1436
Residual 21.9472266 46 0.47711362 R-squared 0.2014
        Adj R-squared 0.0799
Total 27.4814815 53 0.51851852 Root MSE 0.69073
About 20 of the variation of Time used is
explained by the model
41
Survey Findings

• CAPI is more effective than PAPI, CATI or E-mail
  in terms of time
• Women did better than men in filling in the
  questionnaire
• Non-English speaking respondents did better than
  English-speaking respondents
• Familiarity with IT Tools is higher among Asian
  respondents than those from other regions
• Time to fill in the questionnaire is longer with
  non-English speaking respondents through CATI

42
References

• 1. Demographic and Health Surveys-Phase II
  (1990), Institute for Resource Development.
• 2. Designing Surveys, A Guide to Decisions and
  Procedures (1995), Ronald Czaja and Johny Blair.
• 3. U.S. Bureau of the Census. (1992) 1990 Census
  of Population General Population
  CharacteristicsNorth Carolina. Washington, DC
  U.S. Government Printing Office. (1990 CP-1-35).
• 4. SIAP HANDOUTS, 2007
• 5. JICA-TIC Survey 2007 conducted by participants
  of the Fifth Group Training Course in Modules on
  Core Official Statistics

43
Data Dissemination

• CD
• Web
• Hard Copy

44
THANK YOU