Rubbish data

Rubbish data

Through my experience of working with many organizations, research centers, and academic researchers, I have noticed an issue in the collected data that only can be named as rubbish data or useless data.

The idea of useless data can be summarized as data or questions asked in questionnaires that are not useful in anything related to the objectives of the research, for example in many monitoring or evaluation activities, questions are asked in beneficiary interviews about the family structure in detail, such as asking about the family members disaggregated by gender and age groups. Some may think that these data are important, but experience says the opposite, as these data are important in the phase of needs assessment and selection of beneficiaries, which were already collected in the previous activities, and all the cases I witnessed did not use this data (in the course of writing a monitoring or evaluation report), and in the best case, the family members data were grouped into a final number, so why were all these details asked and make the beneficiaries exhausted with all these questions?

The belief of some researchers that if these data are not useful, it will not cause any issues is wrong, as a large number of questions and asking questions that have nothing to do with the research objectives causes several problems, including an increase in costs, an increase in the participants’ hesitation and fear due to a large number of details that are asked about and the lack of Its rationality, the decrease in the participants’ interest in providing serious answers due to the increase in the duration of the interview and their fatigue, an increase in the possibility of errors in data collection, an increase in the complexities of data analysis, distracting the researcher from the processing data and writing the report and thus discussing topics that not related to the objectives of the research and distracting the decision-makers.

The observed cases that may be called rubbish data are uncountable. Asking about the name of the participant in a political poll in which the name of the participant does not matter at all, it only expresses his legal personality as a representative of a sample of the surveyed community groups (except in rare cases related to verification and follow-up of the data collection teams), asking about the participant’s name will necessarily lead to providing answers that stray more from his true opinions, as a result of his fear of linking those answers to his name and exposing him to any harm. I always advise that the questions we ask to be linked to the objectives of our research and not to say, “We wouldn’t lose anything if we ask this question.”

By:
Ghaith Albahr: CEO of INDICATORS