DISCLAIMER: I’m in my 50’s, I’m a dad, a CEO, and a nerd. I have a Ph. D. in applied behavior analysis, and I’m fascinated by the inner working of our minds. My writing will be sprinkled with geek references, dad jokes, phrases that were barely cool when introduced 10 to 20 years ago, brain science bits, and CEO business babble.
King Croesus, the leader of ancient Lydia, was said to be the wealthiest person of his time. The term, “rich as Croesus” lives on today. Could more money be the key to addressing our continued staffing shortage?
Open Shifts, No Shows, and Call Offs– a continuing issue
On a Tuesday in late July, we had 35, eight-hour weekend shifts, and a nine-hour shift, all open—for a total of 289 hours—uncovered, Direct Support Professionals (DSPs) holes in the schedule.
On top of this were another 24.5 hours of partial shifts—bringing the schedule holes up to 313.5 hours.
Then sprinkle in 55 medication passes also not covered on the schedule. Let’s call those 0.5 hours each for another 27.5 hours. With this, we sit at 341 hours (or 8.5 Full Time Equivalents or FTEs).
We can’t forget about call offs. Recall from an earlier blog that we average 16 absences per weekend. This was a payday weekend and our average jumped to 17 absences, adding another 136 hours of openings on the schedule.
In total, in this one weekend, we had 477 hours or 11.9 FTEs open, uncovered, holes in the schedule.
Given the number of open shifts, we decided to try an experiment. What would happen if we paid double time for this weekend? It may sound extreme, but what if the pay incentive resulted in significant improvements in shift attendance? Improvements in attendance would reduce overtime costs. Perhaps saving on overtime costs would allow double time on weekends to be a viable option.
First, we determined our measures:
- Total number of shift call offs.
- The average percentage of the 10 homes that we’d like to have two DSPs working during “awake shifts,” where there are two DSPs. This measure helps us understand if there are potential improvements in the quality of life for people served, as well as better working conditions for DSPs.
- The ratio of call offs to DSPs employed. This helps us understand if attendance improves/does not improve as we add more DSPs to our schedules.
We ended up using an A-B-B2-A experimental design. This means that we started by collecting baseline data on a payday weekend with normal pay (no double time).
We then ran Experiment #1 which consisted of DSPs earning double time for the weekend if they had perfect attendance for the weekend. Experiment #1 produced intriguing results, so we added Experiment #2.
Experiment #2 consisted of paying DSPs double time if they had perfect attendance for the entire week. This experiment was done to see if the offer of double time could lead to improved attendance during the week (Monday through Friday). To make weekend double time practical, we’d need to see significant improvements in attendance and associated reductions in overtime throughout the week.
We then returned to baseline.
Let’s look at shift call offs first. In the table below, we can see that there wasn’t much change in the number of shifts call offs for baseline, Experiment #1, or Experiment #2. We did see a spike in absenteeism when we returned to baseline.
Discouragingly, we saw no improvement in weekday absences in Experiment #2, where DSPs could earn double time if they had perfect attendance for the entire week. The number of shift call offs (29) was the same as the average number of absences we usually experience.
Why did we see a spike in call offs when we returned to baseline? It could be what behavior analysts call a post-reinforcement pause. Basically, after working to gain reinforcement, we tend to take a break before ramping back up. In other words, DSPs may have banked sufficient money during the double time earnings weekend, which made calling off financially easier for them. Or, DSPs may have picked up more hours and been worn out from working those additional hours. These are only guesses and it is possible that more than one variable was at play.
Next let’s look at our quality-of-life measure: the average percentage of homes where we want two DSPs and that had two DSPs working.
The data is more encouraging here. We saw a significant improvement in home coverage during Experiments #1 and #2.
We are hiring a significant number of DSPs and this is a possible, confounding variable. To better understand what effect hiring may have had, we looked at the ratio of shift call offs to DSPs employed.
If this ratio trends downwards, that’s good. Depending on hiring, that would either mean that the number of shift call offs remains the same even as the number of DSP increases, or if we’re not adding new DSPs, then the number of shift call offs are decreasing. In our case, the baseline phases were like the experimental phases. We appear to see improvement in this ratio during the two experimental phases. We did not include shift call offs from DSPs who were not cleared to work alone yet. This may be a confounding variable as new DSPs may be less likely to call off as they just recently started (we need to look at this data).
As with many experiments, there is not a definitive answer to our question of whether paying double time on weekends will improve attendance. Offering weekend double time contingent on perfect attendance during the week did not bend the curve for overall attendance. Seeing improvements in across-the-board attendance is essential if we are to try and make double time on the weekends a regular occurrence. We did see improvements in home coverage, possibly through DSPs picking up shifts during the double time weekends and that is a significant result.
Our Final Assessment
While double time weekends produced improvements in home coverage, it did not result in overall improvements in attendance which is needed to make it financially viable for the long term. Unfortunately, it was not the magic bullet we were hoping it would be. But that is science. Even in failure, we learn something valuable.
Tune in next week to learn “Why the Dinosaurs Were Lucky!”