6.4.1. Data Correction

1. Household size correction: It may be possible that while choosing the random samples, one may choose either larger or smaller than the average size of the population as observed in the census data and correction should be made accordingly.

The first step in data correction is to address household size. The sample collected might not accurately represent the average household size as indicated by census data. For instance, if the census shows that the average household has 4 members, but the sampling results in households with only 2 or more than 6, this mismatch needs to be corrected. This could involve adjusting the data to better reflect the true average household size in the area being studied.

Imagine you are baking cookies, and the recipe calls for 3 cups of flour, but you accidentally use a cup that only holds 2 cups. The cookies will turn out differently they might be too dense or not spread out as expected. Just like you would adjust your recipe to fix the mistake, here we adjust our data to ensure it accurately reflects the reality of household sizes based on census information.

2. Socio-demographic corrections: It is possible that there may be differences between the distribution of the variables sex, age, etc. between the survey, and the population as observed from the census data. This correction is done after the household size correction.

After correcting for household size, the next step involves socio-demographic factors. This means comparing the collected survey data on characteristics like gender and age with the census data. If, for example, the survey has an imbalance in the number of females compared to males, or the age distribution doesn't match the census figures, then adjustments need to be made. This ensures that all demographic groups are accurately represented in the data.

Think of casting roles in a movie. If a film is supposed to reflect a diverse community, but the cast ends up being mostly one demographic group, the film won't resonate with everyone. Ensuring that all socio-demographic groups are represented in the data is like ensuring all communities are represented in a film to tell a comprehensive and accurate story.

3. Non-response correction: It is possible that there may not be a response from many respondents, possible because they are on travel every day. Corrections should be made to accommodate this, after the previous two corrections.

Next, we address non-responses, a common issue in surveys where some individuals do not provide their information. This could happen because they are unavailable due to travel or other commitments. After making the previous corrections, we need to estimate and adjust the data to account for these non-responses to ensure that our findings do not become biased due to missing information.

Imagine trying to gather opinions from a classroom where some students are absent. If you only ask those who are present, you might miss the views of those who dissent because they weren't there. To make sure that everyone's opinion is heard, you could ask those present to provide feedback on what they think their absent classmates would say and adjust your results accordingly. This way, you're attempting to understand the full class's viewpoint, not just those present.

4. Non-reported trip correction: In many surveys, people underestimate the non-mandatory trips, and the actual trips will be much higher than the reported ones. Appropriate correction needs to be applied for this.

The final correction involves accounts for non-reported trips, which are often casual or discretionary trips that survey respondents may forget or underestimate. For instance, trips like going to the grocery store or visiting friends might not be considered important on a survey. To adjust for these missed trips, researchers need to analytically estimate the likely number of trips that should have been reported to better reflect actual travel behavior.

Think of it like filling out a diary of your daily activities. You might remember your important meetings or appointments but forget to jot down the quick stop at the coffee shop or the stroll in the park. If someone asked you how many things you did that day, you might give a lower number than the actual activities. Researchers do the same correction to ensure their travel data is complete and reflects real world behavior.

--