Researching Employee Absenteeism Using the Case Study Method

Hi.My name is Tad Foster, and I'm a Professorat Indiana State University in the Department of HumanResource Development.Today I'm going to talk to you about an applied case studythat I did back a few years ago.I'm going to call it applied because the subject wasa client who came to me as my consulting job--and not as a professor but as a consultant--

to help them deal with the issue that they thought they had,called absenteeism.So during this little presentation here,I'm going to talk to you a little bit about the situation,but then focus in on the design and howdata was collected and analyzed and what we learned.More about, again, the method than

about the actual-- for the company-- what we learned.But it was an interesting little project.Didn't last very long, didn't haveto last very long, to identify what I needed to knowand was able to help the company move forward.So we're dealing with a company that has 32 employees.

This company is an agricultural company.They produce seeds and other things.They do research for other-- for farmersand for other companies.And they have a unique constitution,a unique, if you will, employee base.So the president contacted my department chair--I was a professor at that time at another university--

and they wanted some help.The department chair turned to mebecause he knew I had a consultancyand liked to work in industry and business.And so I went out and met with the president of the company,and he basically said, I have a concernthat I have an absenteeism problem.And in particular, he had a situationwhere his company granted employees the opportunity

to accrue sick leave and vacation time.They were given a certain amount of sick leaveand a certain amount of vacation every month,but they were allowed to bank it if they didn't use it.They were-- that's called accruing it.So they could accrue time.And so he was concerned, because he clearlyhad people using far too much of their available time,

and he wanted some suggestions on howto deal with this problem, as he perceived it.All right, so every employer out thereis concerned about absenteeism and tardiness,but absenteeism and tardiness is essentiallyan issue that really has a single answer.If you truly have an individual who won't report to work,

the fundamental concern is you've gotto discipline that individual.Now, you may take the time to find outif flexible scheduling or some other thingsmight be able to help.But bottom line, employees that don't come to workdon't get paid, right?And so as I went into this situation,it was important for me to figure out,first of all, what I needed to know.

So in meeting with the president, I left that meeting,developed a plan, and sent it back to him and said,this is what I intend to do.So I intend to go in here and tryto get a better understanding of your employees' behaviorand attitudes.And I'm going to try to provide you with the recommendationsthat I think, from a management point of view,would help your company improve this situation.

Now, again, the president had a specific purpose in mind.I gave him this purpose, and he accepted this purpose.So I felt that now what I'm talking aboutis, I've got to go in there and understand the situation.I want the situation to speak to me.So the kinds of questions that I'm

going to frame as my research questions are goingto be guided by, what do I need to getfrom these people from the samplethat I'm going to be looking at?Which is the entire population.Every employee was going to be included.I'm going to look at, what is the policy regardingattendance-- tardiness, absenteeism,sick leave, vacation time?

What is the policy?That is an important thing for me to learn.Second, I need to know, what's the history?What are people doing?Are they coming in late all the time?What's the-- what is going on?The third thing that I need to knowis, how do people feel about the policy and their behavior?Not only how do the employees feel,

but how does management feel?Is there any differences here?So again, you can see that I'm asking ratheropen-ended questions, rather the kinds of questionsthat I don't specify exactly what variables I'm looking for.This is going to be essentially a modified case study.I'm going to do a good deal of data collection,

but it's not necessarily a full-blown, multi-month,multi-dimensional case study.But it is a full case study.So in any case study, modified or otherwise,there are going to be, usually, three pools of data.First of all, interviews.

Second, usually observations, but in this case,observation didn't seem to be appropriate.But there was a pool of data that I desperatelyneeded to get access to, and that was their managementinformation system.What's the data that they've already collected?What's their absenteeism rate and all those kinds of things?How many hours of accrued time did they have?

And then, finally, artifact reviews.All right, so let's take a look at these.And then we're going to talk about this issueof triangulation, because in every case study,we basically are going to triangulate the data.You see, I have three pools of dataput in a form of a triangle to give you a mental model of whatwe're talking about here.The interview data that I collected

was an interview of all 32 employees.And I spent two days in the company.I was able to work through all of them.I didn't have much problem with that whatsoever.People knew I was coming.They were told by the management that theyneeded to talk with me.I kept everybody, and I felt that itwas important procedurally-- whichI'll review in just a moment again--I felt it important for me to say to them,

I'm not going to tell anybody what you said to me,so what you say to me is confidential.I am going to summarize it, but I'm not going to say, oh, yeah,Bill said to me.No, it's going to be an employee or the employees.It'll always be, in summation, data.And I promised them that, because I was concerned thatthey wouldn't talk to me honestly if I didn't.

And so I did clear that up front to let everybodyknow, including the person who washiring me to do this project.Then for those interviews, I set up an interview guide sheet.I needed to know certain things from certain people.And I did have a set of extra questionsthat I was asking management people.On the ex post facto side, I secured the data

from their existing files.Now, today, that would be an electronic system.We call it a management information system.But back then, it was basically Iwent and talked to the clerk who was responsible for keepingthose records.And we pulled the files.And they provided me with all the information.I made a record of those and put them into my computer,

so that I could do the statistics on it.But essentially, they gave me the informationthat I needed for each employee.And again, even as I collected that data, I just simply putemployee number one was a male, was this, was this, and simplyput the data down, did not put any names to it,so I could keep the people confidential,keep the information confidential.

On the artifacts side, what I needed to look atwas the policy manual.I looked around at other things.There wasn't anything on the bulletin boardsthat was essential.What was really essential in this casewas a copy of the employee handbook.And any other policies that they had,whether they be the president wrote a memoto the people and things like that.

So I got all that information.So as I pulled all this information,I was able to then look at what was going on in this situation.And so the procedure was for me to go in, interview,collect all this information, and then pool it all together.

Any time I'm doing interviews, I tendnot to want to do recordings.I tend to not want to spend a lot of time writing notes.I do that afterwards, in what I call a brain dump.I think I can do it.I know I can do it.I've done it over the years.Where I can look at, watch an interview,

talk with an interview-- with an individual,and figure out what they're sayingand what leading questions I need to do or what directionwe need to take the interview and makeit as natural as possible, so that Iget the data that I need.And then I go out of that interviewand write down everything I can remember and fill up my notesand start to look at what I'm learning.

All right.So as we think about this study, I'mcollecting data from three different sources.Now, comes the issue of triangulation.We're going to analyze the data.Now, obviously I'm going to analyze the data differently.First of all, I have got this ex post facto data.Numbers.I know how long the person has been employed.

I know the person's gender.I know the person's classification.And in this company, I found that therewere two classifications: salaried and hourly.The salaried people were all research scientists.Every single one of them was a man.Every single one of them had a Ph.D. On the other side,every single one of them was a woman with a high school

diploma and this was their second job.So over here, we have a career-minded individual,we have a job-minded individual.All right.So I have all of this data.I know how many hours of sick leave they've accrued.I know how many hours of vacation time they've accrued.So I can do the statistical analysis, which in this caseappropriately would be a t-test, because I'm

going to be comparing men to women in one case.And I'm going to be comparing hourly peopleto salaried people, which luckilyfor the amount of statistical review I had to do,was the same.So I had the same number of men as I had salaried.I had the same number of women as I had hourly.

So it was one t-test to determineif there was a statistical differencein the amount of time that they were accruingor the amount of time that they were using.All right.So I've got these three pools of data here.And I'm trying to tell an accurate picture.I want to compare data from each of the pools,

so that in what I'm finding in one,is it being substantiated from whatI'm finding in the other areas?So are my interviews telling me the same picturethat I'm finding from the data?And is it helping me understand the data?Which it is.And then looking at the policy.And I can bring all of these together so

that when I'm writing it up, when I've done my analysisand writing my this is what I foundand this is what I think it means,I have been true to all of the data.And that the three pools of data have substantiated each otherand there aren't conflicting items that would say to me,I need to step back a little bit and get some more data,

because I've got something that doesn't make sense.All right.So again, from a reporting point of view,if you have something that doesn't make sense,this is something that either drops out of the findingsor you find a way to collect the adequate data to answerwhat you don't know.So when you get to the end, if you don't have the data,

you have to tell your reader, I don't know the answer to this.More research is needed.So if we look at the details of whatwe learned during this study-- and I'm focused nowon the content of the study.We only want to spend just a moment here and gothrough these rather quickly.

Essentially, what I found from these three data sources,is that only one person in the entire companyexpressed any concerns.And that was the president.The president was concerned that some of his employeesdid not have an adequate amount of sick leave or vacation timeto protect their salary.

And so this individual wanted some changes.However, the vast majority of the employees-- in fact,every other employee in the company--was satisfied with not only the policy, but the way the policywas being implemented.There were no managers that were complaining about not beingable to get the work done or complaining

about any particular employee.There was no manager who had everfiled a disciplinary action against an employee.Essentially, everyone in the company was happy with this.So we had two distinct groups of people:PhD's that never took time off and hourly people

who took the time off as soon as they got it.They were more concerned about their family.The researchers were more concerned about their career.And so this information clearly supported the other areasthat I was learning.And I came to the conclusion that this company doesn't

have an absenteeism problem.It actually has a policy problem.So when you look at these individuals, on averagethe employees had accrued 620 hours of sick leaveand 140 hours of vacation.They had employees on the Ph.D. side who had maxed out,

who had been with them for 15 to 20 yearsand had maxed out the amount of time that they could accrue.I pointed out to the president of the company thatconstituted a huge financial liabilityand really substantiated my argument that you really herehave a policy issue, not an absenteeism issue.

You start firing people for these reasonsand you will be wrong.This led to some-- for me, as a researcher--and I was still fairly young as a researcher at that time.First of all, it taught me clearlythat the case study approach really worked well.

I was able to draw data from multiple perspectives.And the fact that I didn't know everythingI needed to know when I walked in the doorwasn't a detriment to this study.I was able to pick up all the information that I needed,get access to the information that I needed,and I asked these broader questions that allowed me

the flexibility to look, basically,under any rock that I needed to look under.And so the case study worked extremely well hereand allowed me to substantiate the data that I was receiving.The second thing that I learned wasthat I had an issue of integrity.And this is a question that you're

going to face in your career as a researcher often.Do you remain true to the data?Or do you meet the customer's expectations?In business, the answer is the customer is always right.Well, the customer was wrong.And the data clearly articulated that the customer was wrong.

The customer wanted to correct absenteeismwhen he needed to sit down with whomever, the powers to be,a committee of employees, and redraft the policy statement.The bottom line issue for this presidentwas half of his employees could goto work for any university in the countryand they happened to be in town of one

of a major agricultural research institution.So he had to give them the kind of package that the universitywas willing to give those PhD's.The mistake that the company madewas extending that same package to the hourly employees.Basically, any employee in his companycould take every Friday off and that person couldn't stop them.

They had far too liberal a sick leave and vacation timeto be able to get the work done that they feltthat they needed to get done.So we basically found out that we've gottwo sets of competing values.I value my career.I value a job with a little extra money.And I want to take care of my family first.

Now, what the president and his leadership teamought to focus on is they've got a flexible system.And how can they maintain the flexibility,give the people what they need, and not lose all those hourly?Because if they start changing the policy,those hourly people are going to be unhappy.And they're going to have to work through this slowlyand systematically.

But as a researcher, I'm faced with the issue, do I tell themwhat they don't want to hear?Because that's what the data said.And I've always, in my career, sided with the data.I'm not going to create bad data.I'm not going to tell a story that doesn't exist.I'm not going to cheat.

The data is supreme.What the data told me is what I told them.And then my recommendations are based on the data,not on what I think will make my client happy.