Thank you very much.
Thank you to the committee for having me here today. My name is Geoff Bowlby. I'm the director of special surveys at Statistics Canada. I was the director responsible for the public service employee survey, and that's what we'll be talking about today.
The focus of my opening comments will be on the nature of the survey itself. There are a few numbers thrown in there, but the main purpose is to show you how we collected the information on harassment. I'll also show you a little information, that I'm not sure you were aware we had, on a similar concept of discrimination.
My opening statements come in the form of a presentation. I see you have copies of it. That's great. I have extra copies if some are needed.
What is the public service employee survey, or PSES? The PSES is a survey that provides the opportunity to give feedback on matters that directly affect employees. I'll talk a little bit later about the sorts of questions that were asked on the survey. It was conducted on behalf of the Office of Chief Human Resource Officer about a year ago, between August 29 and September 30.
The date of our release to departments was on January 26, 2012, so it was not that long ago that the information was made available to all departments. The results of the survey are used to measure employees' perception of the state of people management in their organization, to identify strengths and opportunities to guide organizational planning and learning, and to contribute to the assessment of the organization's performance.
Who was surveyed? A lot of people. It was a very large survey, one of the largest that StatsCan undertakes. All employees working in 90 federal departments and agencies were surveyed. It was a census of all of those employees. That means that every single employee in those 90 organizations and agencies were surveyed. The organizations were the ones for which the Treasury Board is the employer, or organizations where the Treasury Board is not the employer but the organization wished to participate anyway. I'll give you an example of one of those organizations in which the Treasury Board is not the employer but wished to participate in any case. Canadian Revenue Agency is a good example of that. Statistics Canada's statistical operations arm is also another example of that. Those are two large parts of the federal public service for which the Treasury Board is not the employer but where the PSES was administered.
It was broadly administered. It's easier to talk about who was excluded from the survey than it is to talk about who was included. Excluded were ministers' exempt staff, employees on leave without pay, and employees on maternity, paternity, or parental leave, because we were targeting people who were at work at the time of the survey. As well, employees from one department who were on secondment to another were not included.
What types of questions were asked? There were a lot of questions asked. There were 98 questions on the current version of the survey. The survey is repeated every three years, by the way. It started back in 1999, so this iteration had some content that was on previous versions of the survey. Of the 98 questions, 79 were the same as in the previous cycles. That's important because if we want to compare over time, it's really only those questions that were consistent from one survey to the next that we can accurately compare over time.
Before the survey was conducted, StatsCan in its role as administrator of the survey took the questions from OCHRO, the Office of Chief Human Resource Officer, and tested those questions in focus groups. We ran a number of focus groups all across the national capital region in both English and in French to ensure that the questions were respondable by the employees we were targeting.
This is a paper version of the questionnaire, but it was rarely administered as a paper questionnaire.
Actually, 95% of the roughly 300,000 people we sent the questionnaire to received it and responded to us by electronic means, so we had e-mail addresses of every employee in the participating departments and we invited employees to respond to an electronic questionnaire by asking them through their e-mail address.
As I said earlier, it's a census of all employees. This is important because in a census there's no such thing as sample error. You are used to hearing poll results, no doubt, where there is a plus or minus 1% 19 times out of 20. That's the sample error. In this case, because it's a census and not a poll or a sample, there is no such thing as sample error.
There are other forms of error that can be introduced. We call that non-sample error. These are the errors that respondents themselves might make. Or Statistics Canada, in the processing of the information, might make a mistake. But those non-response or non-sample errors were kept to a minimum by using that focus group testing process that I described earlier to make sure that there weren't many errors, that respondents understood the questions and could answer them. As well, Statistics Canada was using a tried and proven type of processing of the information to create the database from all these various questionnaires.
The response rate was also very high. Of the roughly 300,000 employees who received the questionnaire, 72%, or 211,000 employees, returned the questionnaire. A 72% response rate on a survey of this type is a very high response rate. It exceeded our target. It was higher than we've ever had in a public service employee survey in the past. It's a higher response rate than any other known government employee survey. The U.K. runs a survey; the United States government runs a survey, and neither of those come anywhere close to a response rate of 72%. So employees were very engaged in the process and responded at a very high rate.
That's an overview of the survey itself. I want to talk to you now about what were the questions related to harassment, what were not the questions related to harassment, and I want to talk to you as well about the questions related to discrimination, which I thought might be of use to the committee, given the similarity in concepts between harassment and discrimination.
I should have said earlier that the harassment and discrimination questions were part of a suite of questions, and if you have the presentation, I would go back to slide 4, where you can see the sorts of questions that were asked. The harassment and discrimination questions were near the end of the survey.
There were four questions related to harassment. Actually, the first is a statement. Because we are asking employees whether or not they perceived harassment, we needed to first define what harassment was for the employee.
The public service employee used the following definition of harassment. This is a definition that was provided to us by the client, by Treasury Board Secretariat:Harassment is any improper conduct by an individual, that is directed at and offensive to another person or persons in the workplace, and that the individual knew or ought reasonably to have known would cause offence or harm. It comprises any objectionable act, comment or display that demeans, belittles, or causes personal humiliation or embarrassment, and any act of intimidation or threat. It includes harassment within the meaning of the Canadian Human Rights Act.
We asked the employees to read that definition before they answered any other questions. It would appear in an electronic questionnaire as a screen of its own. Then, after reading it, the employee would click a button that says “Next” to take them to the first question, which is the broadest question on the concept of harassment, and that is the employee's perception of harassment over the two previous years.
That is our step two. We ask employees to answer if they think they've been harassed: “After having read the definition of harassment, in the past two years, have you been the victim of harassment on the job?”
That is the question that's asked, and there are answer categories of never, once or twice, or more than twice. If you sum once or twice and more than twice, 29% of employees perceived harassment in the two previous years.
That's similar to what we recorded back in 2008 with the public service employee survey. It is the exact same number, in fact, 29%.
By the way, this is the first presentation of any data. You may know that if you go to the Treasury Board Secretariat website, you can see all this information. StatsCan administered the survey, gave the results over to the Treasury Board Secretariat, and they have posted it on their website.
