glevum afghanistan presidential election 2014 wave one survey findings
TRANSCRIPT
1
Core
– Developed
Glevum Associates
Afghanistan Presidential Election
2014 Poll Results (Wave 1)
December 2013
Key Findings from face-to-face survey of 2,148 Afghans in 34 provinces who are likely to vote in the April 5, 2014
Presidential election
CONFIDENTIALITY NOTICE
Unless otherwise indicated by Glevum Associates, the information contained herein is Glevum Proprietary Information and therefore confidential. This brief is intended for the exclusive use of the individual or entity receiving this brief from an authorized Glevum representative. If the reader of this brief is not the intended recipient, you are hereby notified that any dissemination, distribution or copying of this brief is strictly prohibited. If you have received this brief in error or are not sure whether it is confidential, please immediately notify Glevum by e-mail and destroy any copies, electronic, paper or otherwise, which you may have of this brief. The authorized recipient of this brief is requested to protect this proprietary information and not to share this brief with any other person or entity without written authority from Glevum Associates.
Methodology
3
Methodology
4
Methodology - Overview
Glevum Associates is pleased to present the main findings from a national level representative survey measuring the opinions among Afghans in 34 provinces regarding the presidential election due to be held on April 5, 2014.
The study sampled 2,148 Afghan individuals who are likely to vote in the upcoming election and was fielded between November 27 and December 3, 2014 using face-to-face interviews. The sampling margin of error is + 2.11 with a 95% level of confidence. For subgroups, the margin of error is larger.
5
Methodology – Sampling Method Summary
The sample was selected using a randomly-selected, probability sample proportionate to size method. The sample was obtained using the following steps:
1. The number of completed interviews needed in each province was proportionate to the country’s total population. (Population figures from the Central Statistics Organization (http://cso.gov.af/en) were used.) – The number of sampling points needed in each province was determined by
dividing by 10. (An estimated 10 interviews were conducted at each sampling unit.)
2. Each district represented a primary sampling unit (PSU). The number of districts selected in each province was proportionate to the respective provincial population. The districts were randomly selected using “probability proportional to size” (PPS) so that each district or PSU, regardless of population, had the same probability of being sampled.
6
Methodology – Sampling Method Summary 3. Villages in each district were then randomly selected using a simple
random selection process. – Within cities, neighborhoods (nahias) were used; in rural areas towns and
villages were used 4. Starting points (the point at which interviewers begin the household
selection process) were randomly selected using a simple random selection process. – Starting points were recognizable locations – such as mosques, schools,
bazaars, or other easily recognizable locations within each of the selected settlements for the survey.
5. Interviewers used a Kish grid to randomly select individual respondents within each selected household. The Kish grid avoids only heads of household being interviewed.
6. Due to local cultural traditions, the universe was divided at the outset into male and female sub-samples. However, males have registered to vote in larger numbers than females.
7
Methodology – Sampling Method Summary
7. To avoid oversampling females, the samples was split into approximately 60% males and 40% females.
8. Instability and violence in some provinces/districts has required some sampling points to be substituted to keep interviewers out of more unstable and unpredictable areas for their own safety.
8
Methodology - Regions Provinces were assigned to the following regions:
North South East West
Badakshan Daykundi Bamyan Badghis
Baghlan Helmand Ghazni Farah
Balkh Kandahar Kabul Ghor
Faryab Urozgan Kapisa Herat
Jawzjan Zabul Khost Nimroz
Kunduz Kunar Samangan Laghman Sar-e-Pul Logar Takhar Nangarhar Nuristan
Paktika Paktya Panjsher Parwan Wardak
9
The Sample: Achieved Population Sample
Province
Achieved Population Sample
Total Male Female
Achieved Sampling Points 1
Achieved Population Sample Number Percent Number Percent
Badakshan 9 77 44 57.1% 33 42.9%
Badghis 4 36 23 63.9% 13 36.1%
Baghlan 8 84 48 57.1% 36 42.9%
Balkh 12 122 74 60.7% 48 39.3%
Bamyan 4 31 19 61.3% 12 38.7%
Daykundi 4 43 24 55.8% 19 44.2%
Farah 3 43 24 55.8% 19 44.2%
Faryab 9 82 49 59.8% 33 40.2%
Ghazni 11 89 53 59.6% 36 40.4%
Ghor 5 54 34 63.0% 20 37.0%
Helmand 8 71 47 66.2% 24 33.8%
Herat 17 134 80 59.7% 54 40.3%
Jawzjan 5 49 31 63.3% 18 36.7%
1 Average of 9 interviews per sampling point
10
The Sample: Achieved Population Sample
1 Average of 9 interviews per sampling point
Province
Achieved Population Sample
Total Male Female
Achieved Sampling Points 1
Achieved Population
Sample Number Percent Number Percent
Kabul 38 321 192 59.8% 129 40.2%
Kandahar 9 98 57 58.2% 41 41.8%
Kapisa 4 39 22 56.4% 17 43.6%
Khost 5 40 27 67.5% 13 32.5%
Kunar 4 40 26 65.0% 14 35.0%
Kunduz 9 70 39 55.7% 31 44.3%
Laghman 4 41 27 65.9% 14 34.1%
Logar 3 34 19 55.9% 15 44.1%
Nangarhar 14 106 60 56.6% 46 43.4%
Nimroz 1 15 10 66.7% 5 33.3%
Nuristan 1 12 8 66.7% 4 33.3%
Paktia 5 48 28 58.3% 20 41.7%
11
The Sample: Achieved Population Sample
Province
Achieved Population Sample
Total Male Female
Achieved Sampling Points 1
Achieved Population
Sample Number Percent Number Percent
Paktika 4 36 24 66.7% 12 33.3%
Panjshir 1 13 9 69.2% 4 30.8%
Parwan 6 56 36 62.1% 22 37.9%
Samangan 3 26 15 57.7% 11 42.3%
Sar-i-Pul 5 34 19 55.9% 15 44.1%
Takhar 9 87 53 60.9% 34 39.1%
Uruzghan 3 33 29 87.9% 4 12.1%
Wardak 5 56 34 60.7% 22 39.3%
Zabul 3 26 16 61.5% 10 38.5%
TOTALS: 236 2,148 1,300 60.5% 848 39.5%
Note: Males have registered to vote in significantly larger numbers than females. To avoid oversampling females, the samples was split into approximately 60% males and 40% females.
12
Methodology - How “Likely Voters” Were Determined
Summary: To be considered a likely voter, respondents must have had a voter registration card or plan to obtain a registration card and “definitely” or “probably” will vote in the upcoming election.
• Likely voters were determined by asking respondents three questions (provided below). These questions determined whether the respondent had a voter registration card, planned to get a card, and whether he or she said they planned to vote.
• Respondents were asked the following: – Do you currently have a voter registration card?
• If the respondent answered “no, not sure, or prefer not to answer” to the question above, respondents were then asked:
– Do you plan to register to vote so that you can vote in the presidential election in April?
• If a respondent did not have a voter registration card (or was not sure or preferred not to answer) and said that s/he did not plan to register to vote, s/he was not considered a likely voter and was not interviewed.
13
Methodology - How “Likely Voters” Were Determined
• If a respondent had a voter registration card or planned to register to vote, s/he was asked:
– Will you definitely vote for the president in April, probably vote, or will you not vote?
• If a respondent indicated s/he will not vote, s/he was not considered a likely voter and was not interviewed.
• Additionally, if a respondent answered “don’t know/not sure” or “prefer not to answer” to all three questions used to determine likely voters, s/he was not interviewed.
• The requirements for being considered a likely voter will change with waves 2 and 3 by becoming more strict.
Summary of Key Findings
14
Summary of Key Findings
(by Issue)
15
Summary of Key Findings Most Popular Candidates
• Mohammad Ashraf Ghani Ahmadzay has a 4-point lead over Doctor Abdullah Abdullah (29% compared to 25%, respectively). About one in ten likely voters (11%) is undecided.
• Respondents who were undecided or preferred not to say who they would vote for were then asked if there is a candidate for whom they might vote. Of those respondents, 13% said they would vote for Doctor Abdullah Abdullah and 8% said they would vote for Mohammad Ashraf Ghani Ahmadzay.
• When the answers to these two questions – for whom would you vote and, if no candidate is mentioned, for whom might you vote – are combined, Mohammad Ashraf Ghani Ahmadzay remains in the lead with 30% followed by Doctor Abdullah Abdullah at 26%.
• Additionally, when the two questions are combined, 11% are still undecided about which candidate they will vote for.
16
Summary of Key Findings
Most Popular Candidates
• Respondents were asked for whom they would vote, if they could not vote for their preferred candidate. Again, Ahmadzay and Abdullah are neck and neck (21% and 20%, respectively) as the second choice candidate.
• All other candidates are below 10%.
• Support for candidates varies by region but varies only slightly between genders.
• Regardless of whom they support, 26% of respondents still said Mohammad Ashraf Ghani Ahmadzay would win and 21% opined that Doctor Abdullah Abdullah would win.
• Among all candidates, Mohammad Ashraf Ghani Ahmadzay’s support is the most firm with 70% of those respondents who indicated that they will vote for him saying that they will not change their minds before the April election.
17
Summary of Key Findings Voter Preferences
• A majority of respondents (63%) prefer the presidential and vice-presidential candidates be from different tribes or ethnic groups, while a third (34%) says tribal or ethnic affiliation does not matter.
• Fully 84% of respondents indicate that they would vote for a candidate from a different tribe or ethnic group than their own or that it would not matter if the candidate was from a different tribe or ethnic group.
• Most respondents (59%) would or might vote for a female candidate for President. Fully 82% would or might vote for a candidate who asked a woman to run as vice-president.
• Majorities of respondents consider it is important that a candidate listens to all sides of an issue before making decisions; provides specific plans to address AF’s problems; and understand the problems of people like themselves.
• Half (50%) support women having more freedom.
18
Summary of Key Findings • Most respondents indicated that they would not be induced to vote for
an endorsed candidate or such an endorsement would not matter. • Fully 85% said they would not be swayed by President Karzai’s
endorsement or it would not matter. • However, 58 percent said they would consider voting for a candidate
that a family member supports. • Security (49%) tops the list of the most important issues the new
president must address followed by economic issues (17%) and education (6%).
• A sizable majority (61%) would vote for a candidate who wants to negotiate with the Taliban.
• An even larger majority (71%) would vote for a candidate who wants positive relations with the U.S..
• Overall, 40% think it important that candidates should want to keep some international troops in AF after 2014; 51% think it important that candidates want to maintain good relations with Pakistan.
19
Summary of Key Findings
• A large majority (89%) said that they would not vote for a candidate with a history of corruption.
• Similarly, 78% indicated that they would not vote for a candidate with a history of human rights violations.
20
Summary of Key Findings Importance/Fairness of Election • A majority (60%) have heard a lot about the presidential election. Only
8% have heard nothing. • Radio is the most common source of information about the election (47%
get “a lot” of information from it) followed by family members, TV, and friends and neighbors. In contrast, only 2% indicated they get “a lot” of information about the election from religious leaders.
• Respondents resoundingly indicated that it is very important (96%) to elect their leader. Similarly, 86% said that elections are Islamic; only 2% opined that elections are un-Islamic.
• Most respondents are at least somewhat clear about the presidential election process. However a third do not understand the process at all.
• However, most (79%) understand the process for voting on Election Day “very well.”
• Most respondents (55%) opined that it is “very important” that their vote is confidential. And, a majority of respondents (53%) are “very confident” their vote will be confidential.
21
Summary of Key Findings • Most respondents (77%) indicated some confidence that the
election will be fair and transparent. About a quarter (23%) are “very confident” and about half (54%) are a “little confident”.
• Most respondents (65%) are familiar with the Independent Elections Commission (IEC). Of those who are familiar with it, about a quarter (24%) are “very confident” it will manage the elections effectively and fairly; 51% have “a little” confidence.
• Almost half of respondents (47%) are familiar with the Electoral Complaints Commission (ECC). A little over a quarter (28%) are “very confident” that it will resolve complaints fairly.
Concerns About Violence • About half of respondents are concerned about violence in their districts
prior to the election and on Election Day. • Fully 45% are either “very” or “a little” concerned about violence in the
days leading up to the election and 53% are either “very” or “a little” concerned about violence occurring in their districts on Election Day.
22
Summary of Key Findings
• However, only 13% indicated that a threat of violence in their districts would keep them from voting.
• About a third (35%) of respondents are either “very” or “a little” concerned about violence occurring after the election. Most (60%) are not concerned about violence after the election.
23
Main Survey Findings
By Question
24
News About the Elections
A majority (60%) have heard “a lot” about the presidential election. Only 8% have heard “nothing”. However, fewer respondents in the southern region have heard “a lot” about the election than in other regions.
Q1. Have you heard a lot, a little, or nothing about the 2014/1393 presidential election?
60 62
51
61 57
32 33 30
32 32
8 5
19
7 11
0
10
20
30
40
50
60
70
Total North South East West
A lot A little Nothing
25
News About the Elections Radio is the most common source of information about the election (47% get “a lot” of information from it) followed by family members, TV, and friends and neighbors. In contrast, only 2% indicated they get “a lot” of information about the election from religious leaders.
(Asked only of those who have heard “a lot” or “a little” about the 2014/1393 presidential election in Q1) Q2. I’m going to read a list of ways some people get information about elections. For each one, please tell me whether you have heard a lot, a little, or nothing about the election from each item.
47
38
15
41 42
17
32
20
48
29
45
26 27 30
41
22
32
45
2
10
87
2 5
84
1 5
92
0 1
95
0
10
20
30
40
50
60
70
80
90
100
A lot A little Nothing
Radio Family members Television Friends & neighbors Candidate ads
Elders Religious leaders Internet Newpapers SMS
26
News About the Elections Sources of information about the election vary by region. For example, only 7%
of respondents in the southern region get information from television compared to 42% in the eastern region.
(Asked only of those who have heard “a lot” or “a little” about the 2014/1393 presidential election in Q1) Q2. I’m going to read a list of ways some people get information about elections. For each one, please tell me whether you have heard a lot, a little, or nothing about the election from each item.
47 47
63
46 39 38 39
30 37
45
15 13 7
17 16
0 10 20 30 40 50 60 70
Total North South East West
A lot A little Nothing Don't know
42
55
39 34
43 42 36
43 47 36
16 9
18 19 21
0 10 20 30 40 50 60
Total North South East West
A lot A little Nothing Don't know
32 29
7
42
23 20 17 13 20 30
48 53
80
38 47
0
20
40
60
80
100
Total North South East West
A lot A little Nothing Don't know
22 24 16
28
41 32
24 23
34 36 45
50
60
36
23
0
10
20
30
40
50
60
70
Total North South East West
A lot A little Nothing Don't know
Radio Television
Friends and family Tribal elders
27
Most Popular Candidates Mohammad Ashraf Ghani Ahmadzay has a 4-point lead over Doctor Abdullah Abdullah (29% compared to 25%, respectively). All other candidates are below 10%. About one in ten likely voters (11%) is undecided.
Candidate
Q3. Next, if the election for president of Afghanistan were held today, for whom would you vote?
% Mohammad Ashraf Ghani Ahmadzay 29 Doctor Abdullah Abdullah 25 Abdul Qayoom Karzai 8 Professor Abdulrab Rasool Sayaf 6 Zalmay Rasool 6 Abdul Rahim Wardak 5 Mohammad Shafiq Gul Agha Sherzay 4 Engineer Qutbodin Helal 2 Prefer not to answer 2 Mohammad Dawood Sultan Zoy 1 Mohammad Nadir Nayeem 1 Hedayat Amin Arsala * Do not Know/not sure 11 Total 100
*=Less than .5%
28
Most Popular Candidates Respondents who were undecided or preferred not to say who they might vote for were asked if there is a candidate for whom they might vote. Answers to this question indicate respondents who “lean” toward voting for a candidate. When responses to Q3 and Q4 are combined, Mohammad Ashraf Ghani Ahmadzay remains in the lead with 30% followed by Doctor Abdullah Abdullah at 26%.
Candidate
Q3. Next, if the election for president of Afghanistan were held today, for whom would you vote?
[Asked only of those who answered "don’t know" or "prefer not to answer" in Q3] Q4. Is there a candidate you might vote for?
Q3 and Q4 Combined
% % % Mohammad Ashraf Ghani Ahmadzay 29 8 30 Doctor Abdullah Abdullah 25 13 26 Abdul Qayoom Karzai 8 4 9 Professor Abdulrab Rasool Sayaf 6 1 6 Zalmay Rasool 6 6 7 Abdul Rahim Wardak 5 1 5 Mohammad Shafiq Gul Agha Sherzay 4 * 4 Engineer Qutbodin Helal 2 1 3 Mohammad Dawood Sultan Zoy 1 0 1 Mohammad Nadir Nayeem 1 2 1 Hedayat Amin Arsala * 1 * Do not Know/not sure 11 46 6 Prefer not to answer 2 17 2 Total 100 100 100
*=Less than .5%
29
Most Popular Candidates – “Soft” Support More than a third of respondents indicated they might vote for a different candidate than the person they selected in the previous questions (3% said “yes” and 32% said “maybe”) which indicates a level of “soft” support or support that could change. Among candidates, Mohammad Ashraf Ghani Ahmadzay’s support is the most firm with 70% of the respondents who indicated that they will vote for him and saying they will not change their minds before the April election.
(Asked only of those who selected a candidate in Q3 or Q4.) Q5. Do you think you might change your mind and vote for a different candidate in April?
Response choices %
Yes 3
Maybe 32
No 63
Prefer not to answer 2
Total 100
Q5. Do you think you might change your mind and vote for a different candidate in April?
Candidate selected in Q3 or Q4
Yes Maybe No Prefer not to
answer % % % %
Mohammad Ashraf Ghani Ahmadzay 3 24 70 3 Doctor Abdullah Abdullah 5 32 61 2 Abdul Qayoom Karzai 2 29 63 6 Professor Abdulrab Rasool Sayaf 2 47 51 0 Engineer Qutbodin Helal 4 35 61 0 Zalmay Rasool 2 45 50 3 Abdul Rahim Wardak 1 33 66 0 Mohammad Dawood Sultan Zoy 9 51 40 0 Mohammad Shafiq Gul Agha Sherzay 2 40 58 0 Mohammad Nadir Nayeem 7 53 40 0 Hedayat Amin Arsala 0 32 68 0
30
Most Popular Candidates - Gender
Males and females support candidates in similar percentages indicating no discernable gender gap. Females tend to favor the two frontrunners (Ahmadzay and Abdullah) slightly more frequently than males.
Candidate Selected in Q3 or Q4 Gender
Male Female % %
Mohammad Ashraf Ghani Ahmadzay 29 32
Doctor Abdullah Abdullah 25 28 Abdul Qayoom Karzai 9 8 Zalmay Rasool 7 7 Professor Abdulrab Rasool Sayaf 6 6 Abdul Rahim Wardak 6 4
Mohammad Shafiq Gul Agha Sherzay 5 3
Engineer Qutbodin Helal 3 2 Mohammad Dawood Sultan Zoy 1 1 Mohammad Nadir Nayeem 1 * Hedayat Amin Arsala * * Do not Know/not sure 6 7 Prefer not to answer 2 2 Total 100 100
31
Most Popular Candidates - Regions
Support for candidates varies by region. Frontrunner Ahmadzay’s support tends to be located in the northern and eastern regions while, for example, Karzai’s support is largely in the southern region.
Candidate Region
North South East West % % % %
Mohammad Ashraf Ghani Ahmadzay 31 18 36 17
Doctor Abdullah Abdullah 41 15 20 28 Abdul Qayoom Karzai 2 31 8 9 Professor Abdulrab Rasool Sayaf 2 4 5 20 Engineer Qutbodin Helal 1 0 4 3 Zalmay Rasool 9 3 8 3 Abdul Rahim Wardak 3 3 8 2 Mohammad Dawood Sultan Zoy * 0 1 0 Mohammad Shafiq Gul Agha Sherzay * 18 4 1
Mohammad Nadir Nayeem 1 * * 1 Hedayat Amin Arsala 1 0 * 2 Do not Know/not sure 6 5 4 13 Prefer not to answer 3 3 2 1 Total 100 100 100 100
32
Most Popular Candidates – Second Choice Respondents were asked whom they would vote for president, if they could not vote for their preferred candidate. Again, Ahmadzay and Abdullah are neck and neck (21% and 20%, respectively) as the second choice candidate.
Q6. Let’s say your top choice is no longer available to vote for in the elections. Who would be your second choice as president among the remaining candidates?
Candidate %
Mohammad Ashraf Ghani Ahmadzay 21
Doctor Abdullah Abdullah 20 Abdul Qayoom Karzai 9 Zalmay Rasool 9 Abdul Rahim Wardak 8
Professor Abdulrab Rasool Sayaf 7
Mohammad Shafiq Gul Agha Sherzay 5
Engineer Qutbodin Helal 2
Mohammad Dawood Sultan Zoy 1
Mohammad Nadir Nayeem 1 Hedayat Amin Arsala * Do not Know/not sure 13 Prefer not to answer 4 Total 100
33
Voter Preferences – Tribal and Ethnic Affiliations A majority of respondents (63%) prefer the presidential and vice-presidential candidates be from different tribes or ethnic groups. A third (34%) say tribal or ethnic affiliation does not matter.
Q7a. Should the presidential and vice-presidential candidates be from different tribes or ethnic groups, the same tribe or ethnic group, or does it not matter?
63
1
34
2
0
10
20
30
40
50
60
70
Different tribes or ethnic groups
Same tribe or ethnic group
Does not matter Don't know
34
Voter Preferences – Tribal and Ethnic Affiliations
Fully 84% of respondents indicate that they would vote for a candidate from a different tribe or ethnic group than their own or it would not matter if the candidate was from a different tribe or ethnic group. Another 15% said they might vote for a candidate from a different tribe or ethnic group.
Q7b. Would you vote for a candidate for president who is from a tribe or ethnic group that is different from yours, or would it not matter?
57
16
0
27
0
10
20
30
40
50
60
Yes Maybe No Would not matter
35
Voter Preferences – Vote for a Female Candidate
Most respondents (59%) would or might vote for a female candidate for president. Fully 82% would or might vote for a candidate who asked a woman to run as vice-president.
Q7c. Would you ever vote for a female candidate for president? Q7d. If a candidate for president asked a woman to run as a vice president, would you vote for that presidential candidate, or not?
42
17
36
4 1
59
23
16
2
0
10
20
30
40
50
60
70
Yes Maybe No Don't know Prefer not to answer
Q7c. Vote for a female presidential candidate
Q7d. Vote for a candidate who asked a woman to run as vice-president
36
Voter Preferences – Vote for a Female Candidate
Females are much more likely than males to consider voting for a female presidential candidate (23% and 68%, respectively) and to vote for a presidential candidate who asked a female to run as vice-president (46% and 77%, respectively).
Q7c. Would you ever vote for a female candidate for president?
Q7d. If a candidate for president asked a woman to run as a vice president, would you vote for that presidential candidate, or not?
46
29 23
2
77
15
6 2
0
10
20
30
40
50
60
70
80
90
Yes Maybe No Don't know
Male Female
23 24
48
4 1
68
9
19
3 1 0
10
20
30
40
50
60
70
80
Yes Maybe No Don't know Prefer not to answer
Male Female
37
Most Popular Candidates – Who Will Win? Asked whom they think will win the presidential election regardless whom they support, a plurality of respondents (26%) said Ahmadzay will win. However, 3 in 10 (30 percent) are not sure whom will win.
Q8. Regardless of whom you support, who do you think will win the presidential elections scheduled for April 5, 2014?
Candidate %
Mohammad Ashraf Ghani Ahmadzay 26
Doctor Abdullah Abdullah 21
Abdul Qayoom Karzai 6
Zalmay Rasool 4
Professor Abdulrab Rasool Sayaf 4
Abdul Rahim Wardak 3
Mohammad Shafiq Gul Agha Sherzay 2
Engineer Qutbodin Helal 1
Mohammad Dawood Sultan Zoy 1
Mohammad Nadir Nayeem * Hedayat Amin Arsala *
Do not Know/not sure 30
Prefer not to answer 2 Total 100
38
Voter Preferences – Most Important Issues for New President to Address
Security (49%) tops the list of the most important issue the new president must address followed by economic issues (17%) and education (6%).
Q8a. Next, what is the most important issue the next president of Afghanistan must address?
1
1
1
1
1
1
1
1
1
1
2
2
2
3
3
6
6
17
49
0 10 20 30 40 50 60
Don't Know Defeat/remove Taliban
Health care Islamic principles; sharia law
Pakistan Agricultural issues
Develop mining Improve transportation Negotiate with Taliban
Unite AF Remove/punish/monitor foreigners
Solve country's problems Equality; rule of law; human rights; justice
Women's issues Provide basic necessities
Government reform; corruption Improve education
Economic issues Security issues
39
Voter Preferences – Second Most Important Issue for New President to Address
Three in ten (30%) of respondents named economic issues as the second most important issue the new president must address; security is the next most mentioned issue (22%) followed by education (10%).
Q8b. Next, what is the second most important issue the next president of Afghanistan must address? (open ended)
1 1 1 1 1 1 1 1
2 2 2 2 2 2
3 4 4
7 10
22 30
0 5 10 15 20 25 30 35
Don't know Defeat/remove Taliban
Health care Illegal drugs
Islamic principles; sharia law Maintain positive relations with other countries
Pakistan Remove/punish/monitor foreigners
Agricultural issues Develop mining
Improve transportation Negotiate with Taliban
Solve country's problems Women's issues
Unite AF Equality; rule of law; human rights; justice
Provide basic necessities Government reform; corruption
Improve education Security issues
Economic issues
40
Voter Preferences – Candidate Qualities Majorities of respondents want a presidential candidate to have experience in international relations (85%) and have a college education (86%). Only 21% indicate they would like the candidate to be younger than 50 years old, although for most (63%) it does not matter.
85
4 7
4
21 15
63
1
86
2
10
2
0
10
20
30
40
50
60
70
80
90
100
Yes No Does not matter Don't know
Q9a. International relations Be young (less than age 50) Have a college education
Q9. I’d like to ask you about some traits that a presidential candidate may or may not have. Should a presidential candidate [INSERT ITEM], or does it not matter?
41
Voter Preferences – Candidate Qualities Majorities of respondents consider it important that a candidate listens to all sides of an issue before making decisions; provides specific plans to address AF’s problems; and understand the problems of people like you. Half (50%) support women having more freedom.
Q10a-f. Next, is it important, not important, or does it not matter if a presidential candidate [INSERT ITEM]?
50
87
43
46
99
95
30
6
30
36
1
1
18
4
25
14
0
2
2
3
2
3
0
1
0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100%
Supports women having more freedom
Listens to all sides of issues before making decisions
Wants to keep some international troops in AF beyond 2014
Wants good relations with Pakistan
Provides specific plans to address AF's problems
Understands the problems of people like you
Important Not important Does not matter Don't know Prefer not to answer
42
Voter Preferences – Candidate Qualities Women are much more likely than men to indicate that it is important for a presidential candidate to support women having more freedom (36% and 69%, respectively). Men and women tended to see eye-to-eye on other items in this battery of questions.
Q10a. Next, is it important, not important, or does it not matter if a presidential candidate supports women having more freedom such as working outside of the home.
36 40
22
2
69
16 14
1 0
10
20
30
40
50
60
70
80
Important Not important Does not matter Don't know
Male Female
43
Overall, 40% think it important that candidates should want to keep some international troops in AF but by region this varies from 19% in the south compared to 51% in the north. A slight majority (51%) think it important that candidates want good relations with Pakistan. This opinion is highest in the South (68%) and lowest (40%) in the East.
Voter Preferences – Candidate Qualities
Q10c. Next, is it important, not important, or does it not matter if a presidential candidate wants to keep some international troops in Afghanistan beyond 2014?
Q10d. Next, is it important, not important, or does it not matter if a presidential candidate wants good relations with Pakistan?
46 42
68
40
62
36 34
25
43
25
14
21
4
12 10
3 3 2 4 2 1 1 1 1 0
10
20
30
40
50
60
70
80
Total North South East West
Important Not important Does not matter
Don't know Prefer not to answer
43
51
19
44 43
30
20
56
31
25 25 28
23 23
32
2 1 2 2
0
10
20
30
40
50
60
Total North South East West
Important Not important Does not matter Don't know
44
Voter Preferences – Candidate Qualities A large majority (89%) said that they would not vote for a candidate with a history corruption. Similarly, 78% indicated that they would not vote for a candidate with history of human rights violations. Responses to these questions varied by region with, for example, almost a quarter (23%) of northern respondents indicating that they would either vote for a candidate with human rights violations or it would not matter
Q11a. Would you vote for a candidate who has a history of human rights violations, or would it not matter?
4 8
3 2 3
78 70
86 80 81
10 15
3 9 11
5 4 8 6 3 3 3 3 2
0
10
20
30
40
50
60
70
80
90
100
Total North South East West
Yes No Would not matter Don't know Prefer not to answer
Q11c. Would you vote for a candidate who has a history of corruption or bribery, or would it not matter?
4 1
20
3 1
89 91
74
90 93
4 5 4 3 3 2 1 1 2 3 1 2 1 2
0
10
20
30
40
50
60
70
80
90
100
Total North South East West
Yes No Would not matter Don't know Prefer not to answer
45
A sizable majority (61%) would vote for a candidate who wants to negotiate with the Taliban, and an even larger majority (71%) would vote for a candidate who wants positive relations with the U.S..
Voter Preferences – Candidate Qualities
Q11b. Would you vote for a candidate who wants to negotiate with the Taliban, or would it not matter?
61
50
87
64
52
18
28
6
14 17
20 21
7
21
28
1 1 1 3
0
10
20
30
40
50
60
70
80
90
100
Total North South East West
Yes No Would not matter Don't know
Q11d. Would you vote for a candidate who wants positive relations with the United States, or would it not matter?
71
81
57
66
84
10
3
21
13
5
16 14 18 18
11
2 2 4 2 0 0
10
20
30
40
50
60
70
80
90
Total North South East West
Yes No Would not matter Don't know Prefer not to answer
46
Voter Preferences - Endorsements Candidate endorsements would not prompt most respondents to vote for the endorsed candidate or it simply would not matter to them. Fully 85% said they would not be swayed by President Karzai’s endorsement or it would not matter. However, 58 percent said they would consider voting for a candidate that a family member supports.
Q12a-‐e. If [INSERT ITEM] supported a specific presiden>al candidate, would you consider vo>ng for that candidate or who it not maGer who [INSERT ITEM] supported?
7
12
25
30
58
14
20
26
32
21
71
62
47
36
20
6
5
2
2
1
2
1
0
0
0
0 10 20 30 40 50 60 70 80 90 100
President Karzai
Your Mullah
Your Tribal Elders
Your friends
Your family
Yes No Would not matter Don't know Prefer not to answer
47
Importance/Fairness of Elections - Transparency
Most (77%) respondents are at least somewhat confident that the upcoming presidential election would be fair and transparent. About a quarter (23%) are “very confident” that the election will be fair and transparent; more than half (54%) mentioned being “a little confident” in the election’s fairness and transparency.
Q13. Next, are you very confident, a little confident or not confident at all that the elections scheduled for April 5, 2014 will be fair and transparent?
23
54
15
8
0
10
20
30
40
50
60
Very confident A little confident Not confident at all Don't know
48
Importance/Fairness of Elections - IEC Most respondents (65%) are familiar with the Independent Elections Commission (IEC). Men are more likely than women to be familiar with the IEC (71% and 56%, respectively). Of those who are familiar with the IEC, about a quarter (24%) are “very confident” it will manage the elections effectively and fairly; 51% have “a little” confidence.
Q14a. Are you familiar with the Independent Election Commission or IEC?
Yes 65%
No 27%
Don't know 8%
(Asked only of those who answered “yes” in Q14a) Q14b. Are you very confident, a little confident, or not at all confident that the Independent Election Commission or IEC will manage the elections effectively and fairly?
24
51
21
4
0
10
20
30
40
50
60
Very confident A little confident Not at all confident Don't know
49
Importance/Fairness of Elections - IEC About half of respondents (47%) are familiar with the Electoral Complaints Commission (ECC). Similar to the question about familiarity with the IEC, men are more familiar with the ECC than women (52% to 40%, respectively). Of those who are familiar with the ECC, 28% are “very confident” that it will resolve complaints about the election fairly.
Q15a. Are you familiar with the Electoral Complaints Commission or ECC?
Yes 47%
No 44%
Don't know 8%
Prefer not to answer
1%
(Asked only of those who answered “yes” in Q15a). Q15b. Are you very confident, a little confident, or not at all confident that the Electoral Complaints Commission or ECC will fairly resolve complaints regarding the election or electoral process?
28
45
21
6
0
5
10
15
20
25
30
35
40
45
50
Very confident A little confident Not at all confident Don't know
50
Importance/Fairness of Elections – Understanding the Electoral Process
Most respondents are at least somewhat clear about the presidential election process (36% said they understand the process “very well.”). However a third do not understand the process at all. Men are more likely than women to understand the electoral process with 40% of men answering “very well” compared to 30% of women.
Q16. Overall, do you understand the presidential election process very well, a little bit, or not at all?
36
29
33
2
0
5
10
15
20
25
30
35
40
Very well A little bit Not at all Don't know
40
33
26
1
30
25
43
3
0
5
10
15
20
25
30
35
40
45
50
Very well A little bit Not at all Don't know
Male Female
Male/Female Split
51
Importance/Fairness of Elections – Understanding the Electoral Process
Most (79%) respondents do understand the process for voting on Election Day “very well”, although respondents in the southern region were less likely to understand the voting process than those in other regions.
Q17. Do you understand the process for voting on Election Day very well, a little bit, or not at all?
79 85
64
80 76
18 14
28
17 20
3 1
8 3 4
0
10
20
30
40
50
60
70
80
90
Total North South East West
Very well A little bit Not at all
52
Importance/Fairness of Elections – Where to Vote
Most respondents (83%) provided an answer when asked where people in their areas go to vote. Overall, this suggests that most voters know where to vote. If they know where to vote then they are more likely to vote.
Q18. Where do people in your area go to vote?
83 84 82 88
71
85
78
16 14 17
12
28
14 20
0
10
20
30
40
50
60
70
80
90
100
Total Male Female North South East West
Provided an answer Don't know Prefer not to answer
53
Importance/Fairness of Elections – Voting Attitudes
Respondents resoundingly indicated that it is very important (96%) to elect their leader. Similarly, 86% said that elections are Islamic; only 2% opined that elections are un-Islamic. Women are more likely than men to indicate that elections are Islamic (81% and 91%, respectively).
Q19. Do you think it is very important, a little important, or not important at all that Afghans elect their leader?
Very important 96%
A little important 3%
Not important at all 1%
Q20. Some people feel that elections are un-Islamic, while others feel that they are Islamic. What is your opinion on this topic? In general, are elections Islamic or un-Islamic?
86
2 9
3
91
1 6
2
81
3
14
2
0
10
20
30
40
50
60
70
80
90
100
Elections are Islamic Elections are un-Islamic
Don't know Prefer not to answer
Total Male Female
54
Importance/Fairness of Elections – Electoral Confidentiality
Most respondents (55%) opined that it is “very important” that their vote is confidential. And, most respondents (53%) are “very confident” their vote will be confidential.
Q21. No one is supposed to know for whom a person has voted. Is it very important, a little important, or not important at all that no one knows for whom a person has voted?
55
18
23
4
0
10
20
30
40
50
60
Very important A little important Not important at all
Don't know
Q22. Are you very confident, a little confident, or not confident at all that no one will know for whom you voted?
53
36
8
1 2
0
10
20
30
40
50
60
Very confident
A little confident
Not confident at all
Don't know Prefer not to answer
55
Importance/Fairness of Elections – Concerns About Violence
Fully 45% of respondents are either “very” or at least “a little” concerned about violence in the days leading up to the election. Concerns about violence are highest in the south. Similarly, 53% are either “very” or “a little” concerned about violence occurring in their districts on Election Day. Again, concerns are highest in the south.
Q23. Violence can be used to influence people. Are you very concerned, a little concerned, or not concerned at all about violence occurring in your district on the days leading up to the election?
15
7
32
18
8
30 33 31
28 33
54 60
35
54 55
1 0 2 1 4
0
10
20
30
40
50
60
70
Total North South East West
Very concerned A little concerned
Not at all concerned Don't know
Q24. What about on Election Day? Are you very concerned, a little concerned, or not at all concerned about violence occurring in your district on Election Day?
26
14
53
29
17
27 27
18 24
48 46
58
29
46
32
1 1 0 1 3
0
10
20
30
40
50
60
70
Total North South East West
Very concerned A little concerned
Not at all concerned Don't know
56
Importance/Fairness of Elections – Concerns About Violence
Only 13% of respondents indicated that a threat of violence in their districts would keep them from voting on Election Day. However, 22% of respondents in the southern region said a threat of violence would prevent them from voting compared to only 6% in the western region.
Q25. Would a threat of violence in your district on Election Day keep you from voting on Election Day?
13 11
22
14
6
64 69
46
63
74
22 17
30
23 19
1 3 2 0 1 0
10
20
30
40
50
60
70
80
Total North South East West
Yes No Don't know Prefer not to answer
57
Importance/Fairness of Elections – Concerns About Violence
A majority of respondents (60%) are not concerned about violence occurring after the elections. About a third (35%) are. However, 56% of respondents in the western region indicated they are either “very” or “a little” concerned.
Q26. After the election, some people use violence as revenge if they do not like the election results. Are you very concerned, a little concerned, or not at all concerned that violence will occur in your district soon after the election?
8 14
2 3
14
27 26 25 24
42
60 58 62
67
38
5 2
11 6 6
0
10
20
30
40
50
60
70
80
Total North South East West
Very concerned A little concerned Not at all concerned Don't know
58
Importance/Fairness of Elections – Prior Voting Fully 58% of respondents said they voted in the 2009 presidential election. Of those who did not vote in 2009, a plurality said they were too young in 2009 (31%) and another 30% indicated that they had no voter registration card.
Q27. Now, I’d like you to think back to the 2009 presidential election. There are many reasons why people do or do not vote in elections. Did you vote in the 2009/1388 presidential elections or did you not vote?
Yes 58%
No 42%
Q28. What was the main reason you did not vote in the 2009/1388 presidential elections?
1
3
2
3
4
5
10
11
30
31
0 5 10 15 20 25 30 35
Don't know
Other
Forbidden by husband/family
No election/polling place
Prefer not to answer
Did not want to vote
Security/fear of violence
Personal issues/too busy/out of town
No registration card
Too young
59
Importance/Fairness of Elections – Prior Voting An almost identical percentage said they voted in the 2010 Parliamentary election as voted in the 2009 Presidential Elections. This is not an unusual response given that this is a survey of likely voters.
Q29. Did you vote in the 2010 election for members of Parliament?
Yes 41%
No 58%
Don't know 1%
60
Core
– Developed
Demographics
61
Demographics – Gender and Age Fully 57% of respondents were male. A larger percentage of males are in the sample because more males than females are registered to vote. Respondents tended to be younger with 54% under age 30.
54
25
13
6
2
0
10
20
30
40
50
60
17-29 30-39 40-49 50-59 60+
Age
Male, 57
Female, 43
Gender
62
Demographics - Education Some 40% of respondents have never attended school. Only 47% have received any formal schooling (primary/secondary) whatsoever, with just 19% finishing secondary school. Just 1% were educated in a Madrassa.
D5. What is the highest level of education you have completed?
1
1
1
4
6
19
9
7
12
40
0 5 10 15 20 25 30 35 40 45
Home schooled
Finished Madrassa
Finished graduate school
Finished 4-year college
Finished 2-year college
Finished secondary school
Finished some secondary school
Finished primary school
Finished some primary school
Never went to school
63
Demographics – Household Monthly Incomes Fully 28% of respondents have monthly incomes of 10,000 Afghanis or less; 21% have no income.
• Note: An income of 10,000 Afghanis equals about 200 USD.
Less than 5000 Afghanis
11%
5001- 10,000 Afghanis
17%
10001-15000 Afghanis
18% 15001-20000
Afghanis 12%
More than 20000 Afghanis
8%
No income 21%
Don't know 8%
Prefer not to answer 5%
64
Demographics - Occupation Housewife (does not work outside of the home) is the most common occupation (30%). A quarter (25%) are unemployed.
D7. What do you do for a living?
1
5
5
5
5
7
8
9
25
30
0 5 10 15 20 25 30 35
Prefer not to answer
Other
Student
Teacher
NGO worker
Farmer
Work for government
Business owner
Unemployed
Housewife
65
Demographics - Ethnicity
A plurality of respondents (40%) indicated they are Pashtun; 36% said Tajik. Fully 12 ethnicities are represented in the sample.
Ethnic group %
Pashtun 40
Tajik 36
Hazara 10
Uzbek 8
Aymaq 2
Pashaye 1
Arab 1
Nuristani 1
Turkmen 1
Baloch *
Kirgiz *
Kochis *
Other *
66
Demographics - Language
Dari is spoken most frequently among respondents
Language %
Dari 50
Pashto 40
Uzbeki 8
Pashaye 1
Turkmani 1
Balochi *
Nuristani *
67
voteafghanistan.com
For an electronic (soft) copy of this report plus the survey methods report and
supporting survey data please visit: