Indian Prime Ministers: Private Letters and Personality Dataset (1947–Present) (doi:10.7910/DVN/ROCMWH)

View:

Part 1: Document Description
Part 2: Study Description
Part 5: Other Study-Related Materials
Entire Codebook

Indian Prime Ministers: Private Letters and Personality Dataset (1947–Present)

doi:10.7910/DVN/ROCMWH

Harvard Dataverse

2026-01-29

1

Acharya, Mahesh, 2026, "Indian Prime Ministers: Private Letters and Personality Dataset (1947–Present)", https://doi.org/10.7910/DVN/ROCMWH, Harvard Dataverse, V1

Indian Prime Ministers: Private Letters and Personality Dataset (1947–Present)

doi:10.7910/DVN/ROCMWH

Acharya, Mahesh (University of Alabama)

Harvard Dataverse

Acharya, Mahesh

Acharya, Mahesh

2026-01-24

https://doi.org/10.7910/DVN/ROCMWH

Social Sciences, India, Leader, Prime Minister, Personality, Speech, Text, Letter

The texts of the private letters are categorized by prime minister, and are then processed via Mairesse et al. (2007)’s Personality Recognizer application. First, this application groups the words used in those private letters in categories of different linguistic features from two psycholinguistic dictionaries – Linguistic Inquiry and Word Count (LIWC) (Pennebaker et al., 2001) and MRC Psycholinguistic Database (MRCPD) (Coltheart, 1981). The count and the proportion of those words usage are then standardized, and are plugged into the Support Vector Machines for Regression (SMOreg) model to predict the Big Five personality scores. The scores are measured on a scale of 1 to 7 – lower values indicating the prime minister lower on a relevant trait. For example, Indira Gandhi scores lowest in agreeableness among the prime ministers. This finding, to any observer of Indian politics, fits well with her decision to impose martial law or ‘Emergency’ in the country to curb the dissenting views in 1970s. Indian Prime Ministers have a long and established tradition of writing private letters. Some of these letters got released or published as the time passes by, but many are still not available in the public domain. They are however made available in limited amount for research purposes by the relevant agencies such as Nehru Memorial Museum and Library (NMML). Most of these letters I collected are categorized as (top) secret, personal, and (highly) confidential. These letters cover wide range of subjects. From political issues such as internal party disputes and foreign policy disagreements to the mundane discussion on gifts such as perfume and scented hanger, observation how ‘boring’ UN conferences are and the concern about the secret foreign wife and child of a prominent Indian revolutionary leader, these letters reflect the frank, critical, and laudatory views, opinions, or tone. In all, these letters should closely reflect the truer personality of leaders as evidenced from their relatively free choices of words or verbal constructions. Altogether, almost 200 letters spanning approximately 400 pages are analyzed. The letters cover the time period of 1947 till 2014 from first prime minister Jawaharlal Nehru till Manmohan Singh. Private letters of the current Prime Minister Narendra Modi who begins his third term in 2024 are however not available. Personality scores for Modi are measured by analyzing the texts from his interactive monthly radio talk show (another spontaneous form of speech) titled ‘Mann Ki Baat’ which roughly translates as ‘Inner Thoughts’.

This data project was made possible by a grant from the American Political Science Association. The statements made and views expressed are solely the responsibility of the author.

The articles based on this data project are currently in the process of publication. Once they are published, this dataset will be made available to the public.

<a href="http://creativecommons.org/publicdomain/zero/1.0">CC0 1.0</a>

Label:

IndianPM_Letters_BigFive_scores_graph.png

image/png

Label:

Indian_PrimeMinister_Letters_Text_Nehru_to_MMS.zip

application/zip

Label:

IndiaPM_BigFive_letters.csv

text/comma-separated-values

Label:

IndiaPM_personality_scores_letters_detailed.csv

text/comma-separated-values

Label:

Modi_MannKiBaat_RadioTalkShow.txt

text/plain