Skip to main content

This is the first in a series of articles that puts forward some of the fundamentals of HR, and SDy in particular is foundational to making great decisions within your current and future HR roles. So what is SDy? SDy is a statistical method of calculating the performance different in dollar terms between candidates or current role incumbents. As an example, say you’re recruiting for a role and you’ve narrowed it down to two candidates who can both do the job and are a good fit, but you prefer one over the other. You make the job offer and the first choice candidate comes back asking for an additional $10,000 in their salary – SDy can assist you in making this decision.

To calculate SDy you first need an objective measure which is suitable for predicting future job performance. For example in selection, you might use a general mental ability test, a structured interview, or an assessment center, etc – all these assessments when done correctly will provide you with a score for each candidate. Taking this score you need to calculate the standard deviation, which you can do by following these instructions (if these don’t make sense, there are plenty of explanations on the net, simply Google ‘calculating standard deviation’):
In this example we have 15 applicants who we have assessed through a structured interview and have gained 15 assessment scores, you can calculate Standard Deviation either manually, or simply turn to MS Excel and follow these instructions

Firstly enter your 15 assessment scores into Excel, with each number in its own cell:

95, 68, 58, 64, 84, 68, 72, 75, 59, 90, 89, 78, 82, 65, 67,

Then at the bottom of this list, simply type in the formula =STDEV(insert the range here, i.e. =STDEV(A1: A15)), and you will have your standard deviation, in our case the standard deviation is 11.61.

What the standard deviation tells us is by how much members in a group differ from the mean or average. Typically 68% of our sample will fall within one standard deviation of the mean, that is 68% of our sample scored between 62.65 and 85.88 (which is calculated by adding the standard deviation to the mean for the upper score (74.266 + 11.61 = 85.88), and deducting the standard deviation from the mean to gain the lower score (74.266 – 11.61 = 62.65)).

If you’re still with me fantastic, you’re about to understand why I think SDy is so fundamental to making great decisions in HR. SDy is an estimate of employee performance based on 1 standard deviation. It will estimate for us, the performance difference in dollar terms between an average employee and a high performing employ, as determined by their score on an assessment measure. How about all that in plain English? Let's say we have two candidates, one who scored 75 (so pretty close to the mean of 74.266) and one that scored 89 (again pretty close to one standard deviation difference to the score of 75), how much better will the one that scored 89 perform on the job? This is what SDy estimates for us. Calculating the standard deviation is the hard part which can be accomplished with the help of MS Excel, to calculate SDy all we need to do is multiply the annual salary for the role by 40%, and that will give us SDy. For ease of calculation let’s use $100,000 as of the annual salary for the role we’re recruiting for. So SDy = 100,000 * 0.4, which is 40,000. This is pretty exciting, we now know that there is a $40,000 performance difference between these two candidates. This information is incredibly useful, it can help us assess counter offers – how much better do we predict our first choice is compared to our second choice candidate, and can we justify paying them more than we initially had set aside in our budget for this role, or should we go with our second choice candidate?

I should point out at this stage that we don’t have to use just 1 standard deviation, we may, for example, have two candidates that are only 0.5 standard deviations away from each other, or 0.3 standard deviations away from each other. To calculate these values we simply multiple 40,000 by the standard deviation percentage, i.e. 40,000 x 0.5 = 20,000, or 40,000 x 0.3 = 12,000. So if we take again the example of our two top candidates, but in this example, they are only 0.3 standard deviations apart, and our top candidate wants an additional $10,000 in salary, we might re-evaluation and consider that while better the top candidate isn’t that much better than our second choice to justify a salary $10,000 higher than what the second candidate would most likely accept.

SDy is most commonly applied to selection questions and training, for example in selection it can be used to determine if using an external assessment center is worthwhile for the role you’re recruiting for. Alternatively that training program you’ve been asked to look at, will it lift the performance of trainee’s enough to justify its initial cost?

By Brendan Lys

Operating at the intersection of Human Resources and Data Science, I leverage extensive specialist experience within Human Resources, with the methodologies and approaches of Data Science. This focus on the discovery of actionable insights from data, has been applied to areas such as: remuneration & benefits, workforce planning, recruitment, health & safety, diversity, and training. But what does the application of data science to HR challenges and opportunities actually look like. Within an HR framework the data we work with typically comes directly from our HRMIS, an advantage of using data science methodologies is that we can bring in additional data either held within the organization or from external sources - data which is out of reach from a pure HR analytics approach. Consider for example position descriptions, these contain a wealth of data that we typically ignore as its not in a analysis ready format. A side project I'm working on currently (April 2019) is using text mining on job descriptions to provide insights into which job family the position may fit into. The insights of my work have been enjoyed by organizations across a diversity of sectors including: Government (Australia and New Zealand), ASX and NZX listed companies, utilities, not for profit and higher education.