Difference Between Correlation & Causality

What follows in this report is the answers to a series of questions relating to the ARC and their use of performance indexing and measurement to gauge and measure the need for and efficacy of their services. The ARC is responsible for a number of different governmental programs and efforts and they use a number of metrics to measure how well they are doing their job, what the level of need happens to be and so forth (OPM, 2017; ARC, 2017).

There are active questions as to whether the metrics chosen by the ARC as applied and used from the OPM's performance management framework are proper and best for what is being looked at and assessed. Thus, looking at whether ARC is looking at the correct items, whether they are being looked at the right way or if perhaps they should be drilling down a bit further when it comes to the depth and breadth of their analysis is important. The ARC's methodology and choice of analysis points will be compared and contrasted with the overall desired framework and list of best practices from the OPM (OPM, 2017; ARC, 2017).

Answer to Question One: The organization in question, that being the ARC, deigned to look at what they refer to as the distress index of their area. In total, they looked at a total of 410 counties and they used three overall factors, those being three-year unemployment levels, per capita market income levels and decennial poverty rates. The index was necessary because the ARC wanted to get a fix on the amount of distress that the constituents and residents of the area are undergoing. Indeed, the three factors used are all a good indicator of when problems are underway and ongoing. First, unemployment needs to be measured and this is because it is a good indicator of the health of the job force. Looking at individual cases can lead to aberrations and thus looking an aggregate data set using the performance management index is a good way to go. Further, the amount of per capita income is necessary because it gives the average amount per person rather than just looking at the overall pool of money. This part of the index is necessary since it drills down, much like unemployment does, on the proportion of people that are likely or definitely in distress. Indeed, if the overall pool of income is $1,000,000, the amount of distress and problems will be much larger if there are 2,000 people versus 1,000 people. With the former, the amount of income per person is lower and it is higher with the latter one. Thus, it is important to measure the proportion and thus this should be part of the index ARC is using. Finally, looking at poverty rates is important to assess and know. It is necessary to include that in the performance metrics and what is looked at since people below a certain income level, by and large, will be putting up with a lot of distress and economic angst. Thus, this is a necessary and needed part of any distress index on part with what the ARC executed and the model that the GPRA led to them being delivered (OPM, 2017; ARC, 2017).

Answer to Question Two: The employer uses the index to communicate a fairly basic and necessary thing. Indeed, the ARC wants to and needs to get a measure of what is going on in terms of poverty, employment issues and income levels in their area. If one, two or all of those three measures shows that people are in distress, then that will indicate what is needed in terms of support and better outcomes for the people in the area. The results of the ARC's use of the index shows that a good number of the people in the area are in fairly bad shape. There is apparently about 778 distressed counties in the country and 165 of those are within the coverage area and jurisdiction of the ARC. Beyond that, the quartiles that those people inhabit very much trend towards the lower quartiles rather than the higher ones (OPM, 2017; ARC, 2017).

Answer to Question Three: In general terms, the ARC can use the measurements and readouts from the index to determine what level of service is needed as compared to what is currently in use. Indeed, if programs are heavily ramped up but the metrics are good, then the amount of funding and resources deployed can probably be dialed back. However, if the outcomes are more and more negative, then programs and efforts can be ramped up. Of course, the metrics in question could not be used as a way to measure the efficacy of ARC directly. The presence of unemployment, per capita income and poverty rates are mostly a result of broader economic influences and most of those are outside of the complete control of the ARC. Even so, the ARC can use the results as a way to help make decisions and where and how to deploy resources. If one of the perceived aggravators of the distress index is lack of jobs, lack of educational options or something like that, the ARC can do what it can to mitigate those issues. Regardless, the ARC needs to make decisions and choices based on the data that is present and verifiable and that is what the index helps address (OPM, 2017; ARC, 2017).

Answer to Question Four: When it comes to the answer to the fourth question, part of that was started in the last question. Indeed, ARC's strategies are obviously going to be in reaction and response to what is noted and spotted in the metrics. Given that the ARC's goal is to create and facilitate things like economic resources, readying the workforce, helping with infrastructure, fostering the natural and cultural assets and fostering leadership in the area, the ARC obviously has a huge stake and involvement that allows them to greatly help and assist those people in the area. Given all of that, the strategies and programs that ARC creates, draws down and keeps in operation will be in response to what is seen with the income, poverty levels and unemployment of people in the area (OPM, 2017; ARC, 2017).

Answer to Question Five: The organization is going to have at least some problems with the data that was collected. Indeed, per capita income, poverty levels and unemployment are all very complex and complicated paradigms. Meaning, there is going to be a rather hard time in differentiating between correlation and causality. Unemployment, for example, can be caused by a number of things including global economic problems, national economic problems, employers leaving the area or laying off people, educational skills and attainment being too low and beyond. Quite often, it is going to be a combination of many factors, with different items carrying different weight. The best way to combat this is to drill down a little further and look at things like under-employment, educational attainment, the amount of people that have been in poverty for a long time (as compared to a short period such as a few months), people whose income levels have been stagnant for a long time as compared a short time and so forth (OPM, 2017; ARC, 2017).

Answer to Question Five: The best way for the ARC to combat the problem noted in the last question is to drill down a little further and look at things like under-employment, educational attainment, the amount of people that have been in poverty for a long time (as compared to a short period such as a few months), people whose income levels have been stagnant for a long time as compared a short time and so forth. The point is that high-level statistics can guide further analysis but definitive points and analysis cannot be garnered from high-level statistics and metrics alone (OPM, 2017; ARC, 2017).