VISION POLLS


Vision poll conducted between December 2015 and January 2016


FOLLOW THIS LINK FOR A DETAILED REPORT ON THIS VISION POLL


Meanwhile, this METHODOLOGY was considered

 

Participants

Target population

In order to collect data that will best provide the required information, the target population will be all those aged 18 years and above. To estimate the population of the selected age groups, the age structure observed from the Uganda Demographic and Health Survey (UDHS) 2011 was applied on projected population for mid-2015 and Uganda was projected to have 44% of the population aged 18 years and above (eligible voting age). This will be our characteristic of interest. In other words, our required sample size will be calculated according to this proportion. The country will be divided in to 10 geographical regions of interest and probability proportional to size sampling will be used to determine the number of people to be interviewed from each district.

 

Sample population

 

In order to sample the target population, we employed a multi-stage sample technique. The country was divided into 10 geographical regions. Using the preliminary National Census figures for Uganda (NPHC, 2014), each region’s contribution to the final sample size was calculated in accordance to its population of people who are 18 years and above.

 

With the total number of districts in each region known, we determined the representative number of districts for each region. These yielded a total of 43 districts out of the 112 districts in the country. Using probability proportional to size sampling (PPS), the 43 districts were chosen (keeping in mind the number of districts for each region).

 

The number of people to be interviewed in each district was determined based on the population size of each of these districts. Each district contributed proportionately to the sample size of its region

 

With a sample size of 6720 households and 10 households from each village, it meant that we had a total of 672 villages. Because of Uganda’s social setting, rural villages were over-sampled in relation to the urban villages in a 3:1 ratio. Using a list of all villages in per district, we randomly selected the 672 villages.

 

In the last stage of our sampling, households were selected using a methodology mentioned hereunder.

 

Villages that were inaccessible or hard to reach due to several constraints like impassable roads or unfavorable weather conditions were accordingly replaced following the same procedure.

 

Inclusion and exclusion criteria

 

Inclusion Criteria: any adult (a person who is 18 years or older).

Exclusion: Minors (younger than 18).

 

Using the 2014 preliminary census results released by Uganda National Bureau of Statistics for each of the respective sub-regions, we calculated the proportion of people who are 18 years and above.

 

Table 1: Population and sample distribution by sub-region

Sub-region

Population

Population (18+)

Central 1

4,346,044

1,912,259

Central 2

3,716,865

1,635,421

Kampala

1,516,210

667,132

East Central

3,609,484

1,588,173

East

5,671,652

2,495,527

North

3,581,232

1,575,742

Karamoja

988,429

434,909

West Nile

2,474,824

1,088,923

Western

4,626,977

2,035,870

South West

4,312,378

1,897,446

TOTAL

34,844,095

15,331,402

 

Through probability proportional to size, 43 districts will be sampled. Within each district, depending on the number of people to be interviewed, parishes were selected in a ratio of 2:1 for rural to urban parishes. Table 2 shows the number of districts that were selected for each of the sub-regions.

 

Table 2: Number of selected districts per sub-region

Sub-region

Total number of districts

Number of Sampled districts

Sample size

Central 1

12

5

890

Central 2

11

5

710

Kampala

1

1

290

East Central

10

5

600

East

23

7

1,190

North

15

5

680

Karamoja

7

1

190

West Nile

7

3

470

Western

12

4

870

South West

14

7

830

TOTAL

112

43

6,720

 

 

Household procedure

 

Households will be recruited into the survey using the process mentioned hereunder. Upon arrival to the village of interest, an interviewer will be required to identify a physical feature. Using the left-hand rule, the interviewer will identify the first household to be interviewed. The interviewer will then use a skip routine based on that day’s date to determine the subsequent households. Interviewers will approach the household and ask for the household head or a proxy, after which point the study and interviewer will be introduced. Using a random sampling technique known as a Kish grid, the interviewer will determine the person in that household to be interviewed. Informed consent (oral) will be obtained from the respondent. Interviews are expected to take place from within the households themselves in the comfort of the respondent. However, should the respondent feel as though enough privacy may not be obtained from their homes, the interviewer and the respondent may agree on an alternative location.

 

Data Management

 

Data will be collected in hard copy (paper-form questionnaires) and will be entered into a SPSS database. Forms that will not be correctly filled will be returned to the responsible interviewer for review. It is from this database that analysis will be performed on the data.

 






Vision poll between June 12 and June 20


About the poll

The poll was conducted by Vision Group’s research team between June 12 and June 20, covering a sample size of 6,626 households countrywide. 

Pauline Gato, Vision Group’s research manager said the study was conducted in 43 districts in both rural and urban areas. Based on the 2007/2017 UHDS population projections, Uganda was projected to have 44% of the population aged 18 years and above (eligible voting age). 

This was the characteristic of interest. In other words, the required sample size was calculated according to this proportion. 

The country was divided into 10 geographical regions of interest and probability proportional to size sampling was used to determine the number of people to be interviewed from each district.

Several demographics were taken into consideration that is, in terms of gender, the age group — of which only those 18 and above were interviewed, the education level regardless of whether they are literate or not. 

The margin of error was set at ±5%. In order to calculate the required sample size for the study, the country was divided into 10 subregions. Random selection of respondents at household level was done by the use of the Kish grid method to give everybody an opportunity to be chosen for the interview.

If the selected respondent was available, the interview was then considered successful and the interview was conducted. 

Otherwise, if the selected respondent was not at home call-backs were made and if still unavailable or refused, a substitution was made within the same household following the day’s date (up for even date or down for odd date) otherwise the interviewer had to substitute from the next immediate household, but still looking out for a would-be respondent with similar characteristics as the former. 

Households were selected using the left hand rule, which required interviewers to skip four households after every successful household targeting the fifth (5) household on the left. It was also a requirement for a potential respondent to have lived in that area for a period of two or more years.

The Vision Group research department has long experience in carrying out market research and opinion polling.








 
 

 

 



 



 

Presidential Candidates 2016

logo
logo
logo
logo
logo
logo
logo
logo
logo

2016 Election Events

«July 2017»
MonTueWedThuFriSatSun
26 27 28 29 30 01 02
03 04 05 06 07 08 09
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30
31 01 02 03 04 05 06

Presidential Candidates 2011

logo
logo
logo
logo
logo
logo
logo
logo

Twitter Feed