Political families of Uttar Pradesh

Master file Analysis

Document History

Original Publish Date: 10 June, 2020

Updated on: 07 September, 2020

In this project we are using a novel dataset compiled by Rahul Verma. This dataset covers the candidates contested in the different levels of elections held from 1974 to 2019 in the state of Uttar Pradesh. This includes assembly, general and local body elections. This dataset connects the contestants to the political offices to their family members within that universe.

We define dynast politicians as those who are preceded by family members who are currently active in politics or were active in the past. Family is defined as a set of individuals who are bound by proximate ties based on blood or marriage, and this definition includes father, mother, grand parents, siblings and in laws. Active in politics refers to holding an office in an elected body or being a candidate in elections. According to this definition, the head of family or the patriarch is considered to be a non-dynast in the year of entry. Precisely, because of the fact that he did not enjoy any advantage of having family members in politics at that point of time. But, ones a family member enter into the universe of politics, in later years patriarch will be considered as dynast along with the descendants.Additionally, both the first and second member should be entered to the system through assembly or general elections

Summary statistics

AE - GE

This universe includes winners and runner-ups in the general and assembly elections from the year 1977 to 2019.

  • Total observations: 11550
  • Number of unique individuals: 5286
  • Number of unique families : 300

The following section provides the year and election level wise break-up of the observations

Assembly Elections

Year Count
1974 804
1977 806
1980 806
1985 806
1989 806
1991 794
1993 800
1996 804
2002 806
2007 806
2012 806
2017 806

General Elections

Year Count
1977 158
1980 158
1984 158
1989 158
1991 156
1996 158
1998 158
1999 158
2004 158
2009 160
2014 160
2019 160

Levels

Families are spread at multiple office levels. While majority them are present only at the assembly election level, there are a quite a few of them which is spanned across multiple level of office. The following table summarises that -

Table 1: Linkages of families(Category 2)
level count
AE 154
GE-AE 94
AE-LB 19
GE-AE-LB 19
GE 14

Dynast presence in the past elections

This chart depicts the proportion of dynast among the winners of assembly election since 1974.

AE GE election contest

There could four types of election contests:

  • Dynast wins against against a Dynast
    • Dynast wins against a Non-dynast
    • Non-dynast wins against another Non-dynast
    • Non-dynast wins against a dynast

First table shows count of the contests and the second one shows the proportion of contests. First part represents the winner in both the tables headers.

year fam v/s fam fam v/s non-fam non-fam v/s fam non-fam v/s non-fam
1977 1 9 7 465
1980 NA 15 8 459
1984 1 5 1 72
1985 1 11 13 378
1989 2 28 21 431
1991 1 29 23 422
1993 1 26 16 357
1996 6 49 26 400
1998 5 8 9 57
1999 6 15 5 53
2002 14 41 26 322
2004 6 19 7 47
2007 9 50 51 293
2009 6 25 10 39
2012 21 62 46 274
2014 4 17 21 38
2017 12 64 76 251
2019 3 23 16 38
year fam v/s fam fam v/s non-fam non-fam v/s fam non-fam v/s non-fam
1977 0.00 0.02 0.01 0.96
1980 NA 0.03 0.02 0.95
1984 0.01 0.06 0.01 0.91
1985 0.00 0.03 0.03 0.94
1989 0.00 0.06 0.04 0.89
1991 0.00 0.06 0.05 0.89
1993 0.00 0.06 0.04 0.89
1996 0.01 0.10 0.05 0.83
1998 0.06 0.10 0.11 0.72
1999 0.08 0.19 0.06 0.67
2002 0.03 0.10 0.06 0.80
2004 0.08 0.24 0.09 0.59
2007 0.02 0.12 0.13 0.73
2009 0.07 0.31 0.12 0.49
2012 0.05 0.15 0.11 0.68
2014 0.05 0.21 0.26 0.48
2017 0.03 0.16 0.19 0.62
2019 0.04 0.29 0.20 0.48

Elections won

This table shows the count of the unique individuals with regards to the number of elections they have won

Table 2: Break-up of unique individuals wrt to elections won
n_elections_won n_individuals
0 2089
1 1886
2 665
3 329
4 173
5 74
6 35
7 15
8 12
9 5
10 2
14 1

Density graph of the same:

Density graph graph of the number of elections won by families

This density graph represents the same

Elections contested

This table shows the count of the unique individuals with regards to the number of elections they have contested .

Table 3: Break-up of unique individuals wrt to elections contested
n_elections_contested n_individuals
1 2764
2 1061
3 558
4 333
5 255
6 127
7 78
8 54
9 30
10 10
11 5
12 5
13 5
15 1

This is a density of graph of the number of elections contested by families

Life span

This section looks at the life span of families. Life span is calculated by computing the difference in entry year of the family with the. Life span is simply the longitudinal life of a family. This is calculated by simply computing the difference in the entry year of a family with the year in which the last member has contested. The following graph is a density chart of the the life span of the unique families in the dataset.

Retention

This chart depicts the retention of candidates after each election year.

Land distribution

The dataset contains a variable named land which records the land owned by the candidate/family. The information is recorded in bhigas.

Families

This boxplot shows the difference the land owned by family and a non-family entity.

Political Experience

This boxplot shows the difference in the land ownership according to a candidate’s political experience. Political experience is the sum of term duration of every election that they have won.

Educational institutions

The data set has variable which record the candidate’s/ family’s ownership of school/college/both.

Families

This barplot depicts the difference in the ownership of educational institutions among family and non-family entities.

Political Experience

This depicts the ownership of the educational institutions with regards to political experience of the unique individuals.

caste

Depicts the ownership of educational institutions with regards to the caste of the unique candidate.

Industries

The dataset records the information regarding the candidate’s family’s ownership of industries in 12 different categories. For the ease if analysis we clubbed them to 5 categories.

Current category Old category
Petrol Pump Petrol Pump

Others

3

4

5

Families

Ownership of industries among families and non-families

Table 4: Industry ownership
Politician’s identity Average number of ownersip
Family 1.65
Non-family 0.67

rent thick

Ownership of rent thick and non-rent thick industires with regards to the family type.

Politician’s identity rent-type Proportion
Family non-rent-thick 0.39
Family rent-thick 0.61
Non-family non-rent-thick 0.71
Non-family rent-thick 0.29

Ownership of industries among families and non-families with regards to the industry category.

Political experience

Ownership of industries with regards to the candidate’s political experience

Political experience

Family

Political experience composition among families and non-families.

This tables shows the distribution of the unique individuals’ experience/

Experience Categories Count
0 2164
1-5 1562
6-15 932
16+ 244

This table shows the distribution of the experience categories among the families and non-families.

Experience Categories Non-Families Families
0 2148 16
1-5 1520 42
6-15 811 121
16+ 123 121

caste

This shows the caste composition of the candidates with regards to their experience.