Pre-poll survey: Advantage NDA, but alliance could just fall short of majority

No national wave seen in favour of ruling BJP

April 08, 2019 10:46 pm | Updated November 28, 2021 09:54 am IST

The ruling BJP and its NDA seem to enjoy a distinct advantage over its opponents.

The ruling BJP and its NDA seem to enjoy a distinct advantage over its opponents.

As the nation gears up for the first phase of the Lok Sabha election on April 11, the ruling BJP-led National Democratic Alliance (NDA) seems to enjoy a distinct advantage over its opponents. The CSDS-Lokniti Pre Election Survey indicates that the national mood appears to be to give the ruling party/alliance a second term.

Over the last one year, the ruling party appears to have overcome the emerging mood of anti-incumbency and turned the tide in its favour. Each region of the country appears to have its own trend and political direction. As things stand in the last week of March 2019, the NDA appears close to the majority mark, though the BJP itself seems well short of securing a majority on its own. One must however hasten to add, that there does not appear to be any national wave in favour of the ruling party.

The survey shows that both the BJP and the Congress are likely to register a four percentage point increase in their vote share. However, in the case of the BJP, this rise is not accompanied by a corresponding increase in its seat share.

Both BJP and Congress up by 4 percentage points

 May 2014 actual vote share based on alliances back thenPre Poll 2019 Vote estimate (%)
BJP allies7.46
Congress allies3.77
BSP allies3.54


Does not include vote share of SP and RLD here; SP and RLD secured 3.4% and 0.1% votes respectively in 2014.

Note: 2019 estimate based on NES Pre Poll Survey 2019 by CSDS-Lokniti

Despite increase in vote share, BJP likely to lose seats due to a more united Opposition in some key States

 2014 Seats Won based on alliances back then2019 Seat Estimate (Pre Poll)
Congress allies1541-51
BJP allies5341-51

Note: 2019 estimate based on NES Pre Poll Survey 2019 by CSDS-Lokniti

*Does not include SP and RLD; SP won 5 seats in UP which are part of Others in 2014 in this Table.

In some key States of North and Central India, the party is likely to lose seats as it may find it difficult to achieve the high 'strike rate' it secured in 2014.

The SP-BSP combine in Uttar Pradesh is posing a stiff challenge to the BJP. If the current trends are to continue, the BJP is likely to lose a significant chunk of seats in this politically important State. While it may continue to do well in Rajasthan, Madhya Pradesh and even Chhattisgarh, it is unlikely to repeat its 2014 performance here.

The BJP could well make up part of this deficit it suffers in North and West through marginal gains in the East and Northeast. Its tally in the South is unlikely to see any visible improvement. The overall shortfall the party is likely to face could well result in other constituents of the NDA playing a crucial role in government formation.

Pre-Poll Survey-based Seat estimates for major States

StatesUPA (%)NDA (%)Others
Andhra Pradesh0-3


TDP – 6-12

YSRCP – 10-16

Madhya Pradesh6-1217-230-0
Tamil Nadu25-354-140-0
Uttar Pradesh0-632-4038-46**
West Bengal3-72-630-36***

Source: Based on NES Pre Poll Survey 2019 by CSDS-Lokniti

*All for BJD

**All for SP-BSP-RLD Gathbandhan

***All for TMC

Non-UPA parties

It is important to highlight that as things stand today, the projected rise in the NDA vote share appears to be at the cost of the non-UPA opposition. Ironically, among supporters of these non-UPA parties, the sentiment against the NDA is not very strong. Close to one-third of non-UPA/non-BJP voters prefer not to take a stand on the question of the Modi government getting a second chance and just four of every 10 are against giving the ruling party another opportunity.

Possibly, the supporters of the non UPA opposition see a limited opportunity of the parties they support coming to power and this is reflected in their stand on a second chance for the BJP. In any case, this opens possibilities for the BJP in States like West Bengal and Odisha.

If one were to undertake a region-wise analysis, it is clear that the NDA appears to be doing extremely well in North, West and Central India. It is also catching up in the East and North East, though it is faring poorly in the South. While the NDA has made some gains in the South in the last one year, it has been unable to catch up with either the UPA or other regional parties in this region. South India is the only region where close to half the respondents (45%) believe that the country is heading in the wrong direction.

Region-wise voting preference based on Pre-Poll Survey

 UPA (%)NDA (%)Others (%)
West and Central41536

Source: NES Pre Poll Survey 2019 by CSDS-Lokniti

Southern India is the only region that thinks that the country is headed in the wrong direction

 Country headed in the right direction (%)Country headed in the wrong direction (%)No response (%)
All India (19 States)402733
West and Central462331

Source: NES Pre Poll 2019 conducted by CSDS-Lokniti

Question asked: In general, do you think things in the country are headed in the right direction or the wrong direction?

Among social groups

The BJP seems to be ahead of its opponents on account of the support it enjoys among critical social groups. It retains its lead among both upper castes and the OBCs. While it gets more support among the upper and middle income groups, its support among the low income group and the poor is not insignificant.

Young voters (below the age of 25) seem to be more enthusiastic in their support for the BJP. The support for the ruling party declines with age and between the oldest and youngest age groups there is an eight percentage point difference.

The BJP, which a year ago appeared to have lost ground to the Congress in small towns and cities, appears to have made up the deficit and is now ahead of the Congress in villages, towns and both the small and large cities. The farmers, among whom support for the party had declined last year are once again rallying around the BJP.

UPA loses May 2018 advantage over NDA in small cities

 May 2017 (%)Jan 2018 (%)May 2018 (%)Pre Poll 2019 (%)
Vote for NDA46413740
Vote for UPA25292930
Vote for Others29303430
Small cities    
Vote for NDA43353138
Vote for UPA28343828
Vote for Others29313134
Big cities    
Vote for NDA42474449
Vote for UPA37283431
Vote for Others21252220

Note: Big cities are those whose population is above 5 lakh.

Source: Mood of the Nation 2017, Mood of the Nation Jan 2018 and Mood of the Nation May 2018; NES Pre Poll 2019 by CSDS-Lokniti

While close to half the Hindu respondents felt that the BJP should get another chance, more than half the Muslim respondents were against a second chance for the BJP. This assertion of no second chance for the government was much higher among the Christians ( six of every ten) and Sikhs ( seven of every ten).

Religious minorities continue to dislike Modi government; interestingly Sikhs and Christians are slightly more opposed to it than Muslims

 Modi govt. should get another chance (%)Modi govt should not get another chance (%)No response (%)

Source: NES Pre Poll 2019 conducted by CSDS-Lokniti

Above all, besides the social arithmetic, it is political skills that would finally make and unmake the electoral fortunes of different parties. In this respect, a fascinating piece of data deserves mention: When asked about how determined they are to go and vote, three forth of those who support the NDA said they were ‘very likely’ to vote whereas only three fifths of the potential non-NDA voters appeared similarly determined.

Least enthusiastic

Muslims were the least enthusiastic to vote (four of every 10 not sure of voting) and the upper castes were the most enthusiastic (seven of every 10 saying that they would surely vote). In other words, NDA voters are more likely to turn out and vote and non-NDA voters are less likely to turnout to vote.

It is clear that if the UPA and the non-UPA parties do not get their act together, the levels of eagerness among their supporters is not likely to increase as voting day approaches.

Incumbency factor

If the contest is between active pro-incumbency and passive anti-incumbency, that will surely benefit the BJP and its allies. BJP thus, enters the hotly contested election with a clear advantage in more senses than one.

Anti-incumbency vote is getting split between UPA and Others

 Voting for NDA (%)Voting for UPA (%)Voting for Others (%)
Pro-incumbent vote (govt. should get another chance)73819
Anti-incumbent vote (govt. should not getter another chance)115633
No response203248

Source: NES Pre Poll 2019 conducted by CSDS-Lokniti

Sandeep Shastri is the National Coordinator of the Lokniti Network; Suhas Palshikar is Co-director of the Lokniti programme at CSDS; Sanjay Kumar is Director, CSDS and Co-director of Lokniti.


Lokniti-CSDS National Election Study Pre Poll 2019


As part of its National Election Study, Lokniti, Centre for the Study of Developing Societies (CSDS), Delhi, conducted a Pre-Poll Survey between March 24th and March 31st, 2019 among 10,010 respondents spread across 19 States of India - Andhra Pradesh, Assam, Bihar, Chhattisgarh, Delhi, Gujarat, Haryana, Jharkhand, Karnataka, Kerala, Madhya Pradesh, Maharashtra, Odisha, Punjab, Rajasthan, Tamil Nadu, Telangana, Uttar Pradesh, and West Bengal. The survey was conducted in 101 Assembly Constituencies (ACs) spread across 101 Parliamentary Constituencies (PCs). The total sample size targeted was 10,100 with an AC/PC-wise target of 100 interviews. In order to decide the number of PCs to be sampled in each of the 19 States, we first gave special weightage to the small States with 14 seats or less in order to achieve a decent sample of 300 from these States. Thereafter, the number of PCs to be sampled in the remaining medium and large States was determined based on the proportion of their electorate in the total electorate of the remaining States. The resultant target sample of each State was then adjusted to the nearest 100.

The sampling design adopted was multi-stage random sampling. This procedure ensures that the selected sample is fully representative of the cross-section of voters in the country. The PCs where the survey was conducted were randomly selected using the probability proportional to size method (adjusting the probability of choosing a particular constituency according to the size of its electorate). Then, one AC was selected from within each sampled PC using the PPS method again. Thereafter, four polling stations were selected from within each of the sampled ACs using the systematic random sampling method. Finally, 38 respondents were randomly selected using the systematic method from the electoral rolls of the sampled polling stations. Of these 38, we set a target of 25 interviews per polling station.

Once we identified our sample among the electorate, trained field investigators or FIs (a training workshop for them was conducted in each State) were sent to meet them. They were asked to interview only those whose names had been sampled. However, at some locations the non-availability of sampled respondents or difficulty in finding households necessitated replacements/substitutions. Our investigators sat down in the homes of people and asked them a detailed set of questions which could take up to 30 minutes. The questionnaire we presented to our sample of voters was designed in the language mainly spoken in the respondents’ State. In Gujarat in Gujarati, in Kerala in Malayalam, etc. Each PC/AC was covered by a team of two FIs, except Tamil Nadu where a team of four FIs was sent to each AC due to late start of fieldwork. A total 216 field investigators conducted the survey at 404 locations.

The achieved national sample is broadly representative of India’s population, in terms of the country's general demographic profile. Data of each State has been weighted by gender, locality, caste group and religion as per Census 2011 percentages.

Profile of the achieved national sample


Raw share in the achieved survey sample (%)


Actual share in total population of 19 States as per Census 2011 (%)


Actual share in India’s total population as per Census 2011 (%)



















































Note: Figures have been round off.

The survey was coordinated by scholars from the Lokniti Network: E Venkatesu and Srinivas Rao Gangiredla (Andhra Pradesh), Dhruba Pratim Sharma and Nurul Hassan (Assam), Rakesh Ranjan (Bihar), Lakhan Choudhary (Chhattisgarh), Biswajit Mohanty and (Delhi), Bhanu Parmar (Gujarat), Kushal Pal and Anita Agarwal (Haryana), Harishwar Dayal and Amit Kumar (Jharkhand), Veenadevi and K L Nagesh (Karnataka), Sajad Ibrahim and Rincy Mathew (Kerala), Yatindra Singh Sisodia and Ashish Bhatt (Madhya Pradesh), Nitin Birmal (Maharashtra), Gyanaranjan Swain and Balaram Pradhan (Odisha), Ashutosh Kumar and Hardeep Kaur (Punjab), Sanjay Lodha and Nidhi Seth (Rajasthan), P Ramajayam (Tamil Nadu), Vageeshan Harathi and Ramya C. (Telangana), Mirza Asmer Beg, Shashikant Pandey and Sudhir Khare (Uttar Pradesh), and Suprio Basu and Jyotiprasad Chatterjee (West Bengal).

The survey was designed and analysed by a team of researchers at Lokniti, CSDS. The team included Amrit Negi, Amrit Pandey, Anurag Jain, Dhananjay Kumar Singh, Himanshu Bhattacharya, Jyoti Mishra, Manjesh Rana, Sakshi Khemani, Shreyas Sardesai and Vibha Attri. The survey was directed by Prof. Sanjay Kumar, Prof. Suhas Palshikar and Prof. Sandeep Shastri of Lokniti..

0 / 0
Sign in to unlock member-only benefits!
  • Access 10 free stories every month
  • Save stories to read later
  • Access to comment on every story
  • Sign-up/manage your newsletter subscriptions with a single click
  • Get notified by email for early access to discounts & offers on our products
Sign in


Comments have to be in English, and in full sentences. They cannot be abusive or personal. Please abide by our community guidelines for posting your comments.

We have migrated to a new commenting platform. If you are already a registered user of The Hindu and logged in, you may continue to engage with our articles. If you do not have an account please register and login to post comments. Users can access their older comments by logging into their accounts on Vuukle.