Every time a new economic plan is announced by the government, newspapers come out with surveys on how people in general feel about it, and if it's favorable or not for the economy and likewise. In these surveys, they take a cross section of people to ask them the relevant questions and publish their data.
The resulting data gives us information like, sixty percent people are happy and the rest express their disappointment and so on. This data is called the cross sectional data.
It is primarily the observation and description of various subjects researched and applied for the purpose of planning and decision-making. This method of interpretation is common in financial markets, health surveys, sociological research, and the psychological field.
There is a basic difference between the terms 'census' and 'survey'. In a census, every person in a given country is counted, whereas in a survey, a sample is taken which consists of people who are perceived to be representatives of the entire population. Extrapolation is the next step which acts as an extension to represent the entire population.
The data collected is like taking a snapshot of the concerned population or subjects. Census can be defined as the largest cross sectional research with the population being understood at a single point in time.
The method here is, eliminating the time dimensions of the data and making it one-dimensional. This data helps in simplifying the comparison of various entities. It is best expressed with the help of bar graphs and pie charts.
In longitudinal research, we can learn about the differences in the entities over a period of time. For instance, a dataset with the marks of a student over four grades give us longitudinal data, helping us compare his performances over four years.
It can help assess whether there's an improvement or fall in his academic achievements. This type of data judges the long-term phenomena suggesting the change as opposed to its counterpart, which gives only the short-term results.
Various cross sectional data collection techniques consist of questionnaires, interviews and online surveys. The best example of obtaining the data is prediction of the future president every four years or even the predictions of presidential nominees for a specific political party. The method employed is simple.
A random set of people are taken in every state and their predictions are asked and registered. This is done at the same time in all the states to compare people's inclination at that specific instant. The data collected across all the states forms the cross sectional dataset, helping us to compare whose chances are more in various parts of the country.
Some examples are the health surveys conducted by the Centers for Disease Control and Prevention (CDC) to find out the prevalence of diseases and symptoms, or the General Social Survey (GSS) conducted by the National Opinion Research Center at the University of Chicago, to collect data on demographic characteristics and attitudes of the residents.
It is used to correlate many different demographic factors like age, race, gender, as also their belief system and opinions about various matters of national importance.
- Data collection is done swiftly as it eliminates the time dimension. There's no delay in publishing the results and conclusions.
- It is cost-effective as compared to the longitudinal data.
- There's no need to keep track of the entire population over a period of time, as in longitudinal research, which certainly minimizes the effort.
- Lacks the detailed analysis of longitudinal research as it gives us only the differences but fails to show the differences over a period of time.
- The biggest drawback is that sometimes people taken for the research may not represent the entire population. So, the results can be deceiving if seen in the larger sense, which is termed as an ecological fallacy.
- Unplanned or sudden changes in the area of research cannot be taken into account, which can have a lasting impact on the entire research.
- As it eliminates the past and future comparisons, the causes and effects about the topic are unknown, hence failing to answer the precise question.
Overall, we can see that this type of data gives an entire idea of the subject at a specific point in time, but fails in helping to compare the data over the years.
However, the positive factor being, the data can be collected from different strata of society, differentiated into males and females, various age groups, and socioeconomic classes, thus helping us to understand the patterns of thoughts and opinions in a well-defined manner.