The 1994 Census bureau database provides data on 14 variables including income level for 32 561 individuals.
The 14 variables are the following:
- Workclass: Federal-gov, Local-gov, Never-worked, Private, State-gov, Self-emp-not-inc (self employed not incorporated), Self-emp-inc.
- Education: 10th, 11th, 12th, 1st-4th, 5th-6th, 7th-8th, 9th, Assoc-acdm, Assoc-voc, Bachelors, Doctorate, HS-grad, Masters, Preschool, Prof-school, Some-colleg
- Education_num: number of years of education (1-16)
- Marital status: Divorced, Married-AF-spouse, Married-civ-spouse, Married-spouse-absent, Never-married, Separated, Widowed
- Occupation: Adm-clerical, Craft-repair, Exec-managerial, Farming-fishing, Handlers-cleaners, Machine-op-inspct, Other-service, Priv-house-serv, Prof-specialty, Protective-serv, Sales, Tech-support, Transport-moving
- Relationship: Husband, Not-in-family, Other-relative, Own-child, Unmarried, Wife
- Race: Amer-Indian-Eskimo, Asian-Pac-Islander, Black, Other, White
- Sex: female, male
- Capital-gain: 0 - 99 999
- Capital-loss: 0 - 4 356
- Hours -per-week: 1 - 99
- Native country: 40 countries (Cambodia, Canada, China, Columbia,...)
Age : 17 - 90
Here is what the dataset looks like:
By uploading the dataset on Visao platform, the dataset is processed and a graph is created displaying the relations between the values of the 14 variables. The graph is built from the correlation between the variables (a link is displayed between variable values if its lift is > 1,1). The thickness of the link is proportional to the correlation.
The graph above shows that people who have income > 50K$/year tend to with respect to studies/occupation, by decreasing order, be self-employed-incorporated, have executive/managerial positions, be professors, have 12-16 years of education, and to a lesser degree, work as sales, for federal government or state government.
Conversely people who have income < 50K$/year tend to, by decreasing order, have received a lower education (1-9 years of education), be private house servants, handlers/cleaners, machine operators, and undetermined workclass.
People who have income > 50K$/year with respect to family status tend to, by decreasing order, be wife, husbands, married civil spouse, husband and to a lesser degree, work more (top quantile : 41-99 hours/week), rather old (aged 45-90, 32-44).
Conversely people who have income < 50K$/year are the ones who, by decreasing order, are own child, never married, not in family, female, divorced, separated, unmarried, aged (17-31), widowed, young (aged 17-31), work less (1-39 hours/week).
People who have income > 50K$/year with respect to culture tend to, by decreasing order, be from Canada, and to a lesser extent, Asian-Pacific islanders, male, and from undetermined countries.
Conversely people who have income < 50K$/year tend to, by decreasing order, be from Haiti, El-Salvador, Poland, Mexico, Puerto Rico, Cuba, be female, black.
The graph above shows that people who have income > 50K$/year tend to be in the top 3 quintiles for capital gains and capitol loss, whereas those who have income > 50K$/year tend to be in the 2nd quantile for capital gain (114-3411$).
So far we have only tried to relate the income variable to other explanatory variables. But there are many other combinations. One consists in linking education and number of years of education. For someone who does not know what "Assoc-voc" means, the graph provides an additional information: it requires 11 years of education.
If we look at the relationship between education and country of origin, the graph shows that people with Asia/Pacific race tend to have higher degrees (masters and bachelors) but not individual countries (China, India,...), and people from Canada are more bachelors.
We could thus continue to provide more perspectives in this analysis but it would be more useful to do it with people looking at understanding a population structure. Do not hesitate to contact us to deepen this analysis, we will explain the various possible options for the valorization of the dataset, the theoretical foundations of the various indices that Visao application calculates.
