Page 140 - Data Science class 11
P. 140
1. In trial assessments, the subject is assigned to a random treatment and control group. However, it is unethical to
expose the subject to arbitrary treatment in specific scenarios. Thus, observational studies are preferred over trial
assessments. For example, purposefully exposing a subject to polluted air to observe the health issues that come
to the forefront is unethical.
2. Some of the trial assessments may require large sums of money or time to execute. There may be occasions
when such large sums of money cannot be arranged. In such scenarios, it will be a better idea to drop the idea of
performing trial assessments, and give the observational study a priority.
3. A trial assessment cannot be performed in some scenarios, as it becomes unfeasible to assign a subject to a group
randomly.
Qualitative Observations vs Quantitative Observations
• Qualitative observations: These observations are made through the senses to observe the results, i.e., through
sight, smell, touch, taste, and hearing. Observation skills are required to make good use of all five senses to recognise,
analyse, and recall your surroundings. This practice is often associated with mindfulness because it motivates you to
be present and aware of the minute details of your day-to-day life.
• Quantitative observations: These observations are made via measuring instruments like rulers, balances, graduated
cylinders, beakers, thermometers, etc. Since these observations make use of measuring instruments, the results are
measurable.
Observational data is gathered by observing a behaviour or activity. It is collected through methods like human
observation, open-ended surveys, or the use of an instrument or sensor to monitor and record information, as in the
case of quantitative observation. For example, sensors are used to observe noise levels. Observational studies are
ones where researchers observe the impact of a risk factor, diagnostic test, treatment, or other intervention without
trying to change who is or isn’t exposed to it. There are two types of observational studies: cohort studies and case-
control studies.
Observational studies are best used to analyse the real-world applicability of proofs derived generally via randomised
trials and to examine patients and conditions not typically included or studied in randomised trials. This is done to
better understand current treatment practices and how patients are assessed in order.
When to use Observational Data?
Observational data is best used when one of the following situations occurs:
• While collecting sensitive information, when you don’t trust your participants to be truthful with their self-reporting.
• When you need to understand the how or what of a research question.
• When you need robust data to describe consumer habits.
• When behaviour in a natural setting is important to your research question .
• When behaviour in a controlled setting is critical to your research question.
• When you are concerned that self-reported data about behaviours will not be the same as actual actions, despite
being unintentional.
• When you need more information about a particular research question to form a complete and error-free survey in
all respects.
If you are in any of the above stated research states, then you might need an observational study.
138 Touchpad Data Science-XI

