Reflective Project

1 Reflection

Reflection

Evette Grayson

University of Arizona at Global Campus

Dr. Earl Greenia

MHA 605 Business Intelligence

October 10, 2022

2 Reflection

I have learned that a significant part of business intelligence is analysis which is a process

by which data is converted into information. Analysis involves the application of statistics and

experimental design. In the latter, analytic tools are used to conduct experiments on both processes

and the output of processes. Experiments are conducted to check the validity of hypotheses by

showing the real cause and effect by depicting what outcome occurs after a certain factor’s

manipulation. Experimental design is significant in establishing better targets by determining

possible outcomes from a certain process with a particular level of confidence. It also allows for the

implementation of rules, parameters, and recommendations to help in decision-making in real-time

dynamics to accomplish expected results in a situation. Lastly, experimental design allows higher

speed and precision in the decision-making process.

I also learned about sampling or the collection of a data set that is used to make inferences

about a target population. I learned that samples should be representative of the population or

contain all the population’s characteristics otherwise conclusions cannot be drawn because they

would differ from those of the entire population. The two main types of sampling risks are the risk

of incorrect rejection where the sample can produce a conclusion that rejects a theory about the

whole population even though the theory is true about the population or the risk of incorrect

acceptance of the hypothesis where the conclusion can support a theory that does not hold true in

the population. Additionally, the practicability of the sampling should be taken into consideration

such as the availability of the participants or samples, tools, and equipment, ethical considerations,

etc. Moreover, modeling methods are used to determine sample sizes and each method has its own

sampling rule.

3 Reflection

The most common mistakes in analysis are inclusive of sophisticated compensates, isolating

and explaining meaningful patterns, and correlation versus causation. The availability of

applications can tempt an analyst to compensate for a lack of business understanding or data with

sophisticated statistics. This can lead to incorrect techniques for problem-solving and analytics.

Besides, it is common to take noise and randomness in the system into consideration which

demonstrates the complexity of isolating and explaining meaningful patterns exhibited in data.

Lastly, correlation does not necessarily mean that an independent variable is a cause or driver. To

determine causation, an analyst should come up with several business hypotheses for the results

being studied. Additional mistakes include using the wrong tools, not knowing the need for

experiments, not considering system dynamics, and not comprehending the business.

I also learned about the process of opportunity identification and selection. This process

starts by identifying a high-value analytic activity to get involved in and it is followed by creating

and following a work prioritization process. During the entire process, the BI/Analytics consultant

should guide the system users and control and handle analysis requests. He should also meet with

the relevant individuals to strategize and form budgets.

I will use the above information in my future career to ensure a more productive analysis

process as a BI/Analytics consultant. I have grasped the importance of experiments and I have

identified the most common mistakes in analysis that I can avoid in the future. I have learned that

business intelligence is a complex process that requires a lot of knowledge, attention, and

engagement from other system users to gain the best insights.

4 Reflection