The combination of data from multiple unclean sources produces unreliable insights. Begin by locating main data inconsistencies which include duplicated records and missing information together with formatting problems. The data entry standards need standardization while automation tools need to apply cleaning processes to existing records. Clear protocols should be established with all researchers to stop future data errors from occurring. Generating shared survey templates solved formatting inconsistencies from past projects. Any team that addresses bad data early minimizes wasted time while achieving dependable results. The accuracy of research directives depends entirely on consistent data acquisition methods.

When we have different data for the same entity or property but the data are collected by different persons and/or different equipment and/or different time, there always be difference among the data due to randomness or variations. Things to consider with regards to this situation are: - Make sure all data has their estimated uncertainty. Without uncertainty, we cannot reliable compare different data. With uncertainty, we can say whether the data are statistically similar or not. - Make sure all instruments and procedures are calibrated and verified. Random procedures and not-maintained instruments will results in random data and hence lager variation among data. - Make sure that all data has the same reference/unit/scaling.

Inconsistencies are healthier in a research environment. ? It enables self-assessment of the defined research methodology effectiveness and bring areas where clarity/change required. ? It enables each researcher to learn from other’s perspective on why the inconsistencies happening despite all following the defined research methods ? It enables understanding the research impact on different or beyond-the-defined scenario that the result may behave differently