Article 1:
Survey Effectiveness
Index Construct

Article 2:
The PDi Survey Process

Article 3:
Taking Surveys to the Next Level

Article 4:
Customer Loyalty

by Robert A. Page, Jr., Ph.D., and Edward Tamson, Ph.D.

Contact PDi about these publications

(copyright 2002 by PDi. All rights reserved. Please do not modify, copy or distribute this article -- see website terms and conditions of use.)

Executive Summary

The relationship between communication, leadership, and employee satisfaction is well documented. Many organizational educational/training programs, change interventions and business book best sellers are built upon the premise that empowering employees with accurate information and understanding improves employee satisfaction, loyalty and effectiveness. Not surprisingly, the results of this Performance Dimensions International LLC (PDi) benefits project this study confirm those findings. However, one set of results were surprising, namely, that ongoing accurate information actually improves employee satisfaction with company benefits as well. In the PDi employee benefits survey at CAM Systems (the name has been changed to honor confidentiality agreements), employees who rated themselves as well-informed on their benefits gave significantly higher favorable ratings to the benefits packages, despite the fact that each benefits package had readily identifiable areas of weakness.

The question becomes, why do informed employees perceive benefits programs more favorably than uniformed employees? Informed employees are more aware of any problems with benefits, and this could be expected to undermine, not enhance, the favorability of their ratings. Ignorance, on the other hand, is supposed to be associated with bliss, according to the popular adage, in this case translating into more favorable ratings. This was not the case, and there are two possible explanations: informing employees of all the strengths and limitations of company benefit builds openness and trust, while removing uncertainty reduces negative bias.

Openness & Trust Norms. CAM Systems has developed norms of openness and trust with its employees. In our November 2001 organizational effectiveness survey, 1977 respondents gave favorable rating to questions on honesty and openness. The questions were assessed by a 7 point agreement scale, with a non-response option:

2001 Survey Questions	Mean	Standard Deviation
My manager is open and honest with me.	5.9	1.28
The people I work with tend to be honest and open with each other.	5.7	1.30

Given a company atmosphere of openness, employees who feel well informed about their benefits may trust CAM Systems to provide them with fair and reasonable value, particularly in comparison with benefits packages at comparable local firms. Thus, they would tend to rate benefits as meeting or exceeding their expectations, because their expectations of fair treatment have been met, even if their satisfaction with certain features was lacking.

Removing Uncertainty. Second, perhaps removing uncertainty is, in and of itself, an important factor in satisfaction. Employees with knowledge have better information for comparison with other firms, and thus can appreciate CAM System's benefits strategy of exceeding industry norms with the packages they offer. Similarly, knowledgeable employees have less uncertainty about exactly what is and is not covered. It appears that in absence of accurate information, employees assume a worst-case scenario concerning benefit coverage. This negative bias is often reinforced by the vicarious experiences of other employees. Those experiences which are publicly shared usually focus on benefits disasters, not the positive successes - there is little drama in recounting a story where everything goes right. Given a benefits strategy which is designed to match or exceed industry norms, information is power. Without accurate information, no benefit of a doubt is given, and no accurate comparisons with similar companies can be made.

In conclusion, sharing information about benefits programs, even when that information reveals limitations, seems to be a very good idea. Sharing accurate information, warts and all, builds trust and reduces uncertainty, improving favorable expectations of the program, and favorable ratings of the benefits packages.

Methodology

The CAM Systems Benefits Survey was administered by PDi in July of 2002. Two email invitations were sent out to 3855 CAM Systems employee addresses, of which 76 were not viable, due to recent terminations. 1192 employees submitted usable responses through the PDi internet based survey. This represents a response rate of 31.5%.

The Pre-Test. Survey items focused on overall expectations, benefits information, and benefits processes. These questions were rated by a 7 point interval scale where 1 = strongly disagree and 7 = strongly agree, with a non-response option of "Don't Know." Two categorical questions assessed the importance of benefits packages in selection decisions. Open-ended questions for each survey category asked respondents why they gave any of the benefits particularly high or low ratings. Prior to administering the July 2002 survey PDi conducted a survey pretest in June of 2002 on a stratified sample of 150 CAM Systems employees, of which 65 submitted completed internet surveys. Items were examined for face, content and construct validity. As a consequence of the statistical testing, one outcome variable on employee recommendation of job opportunities to friends and family was dropped, due to problems with discriminant validity.

Survey Outcome Variables. The two survey outcome variables assessed employee appreciation and a reputation for competitive benefits, and are single item indicators:
1. Our benefits programs and policies make me feel valued and appreciated as an employee.
2. Compared to similar companies, CAM Systems has an excellent benefits plan.

Agreement Variables. The other agreement scale items were collapsed into categories established using exploratory factor analysis. Only one item double loaded, and was dropped. These categories were retained only if they were reliable, with Cronbach Alpha reliability coefficients over .70.

Benefits Information Category Questions:
1. The on-line enrollment process is easy to use.
2. On-line information on the benefits website is clear and understandable.
3. The information I received during open enrollment was useful.
4. CAM Systems has given me all the information I need to understand the benefits.
5. If I have questions about benefits, I know where to go to find accurate answers.

Medical Benefits Expectations Category Questions:
1. Medical benefits meet or exceed my expectations.
2. Claims/reimbursement processes are simple & easy; I do not feel frustrated/discouraged.
3. My out of pocket costs for these benefits are very reasonable.

Non-medical Benefits Expectations Category Questions:
1. Life and disability insurance policies meet or exceed my expectations.
2.The 401(k) savings program meets or exceeds my expectations.
3.Vacation benefits meet or exceed my expectations.
4. Personal time and personal leave policies meet or exceed my expectations.
5. Work life services (Legal, Domestic Partner Benefits, Fitness Programs, etc.) meet or exceed my expectations.

Survey Findings

CAM Systems employees rate benefits packages as being very important in the two categorical questions. 55% of the respondents indicated that benefits packages were as important as or more important than all other factors in their decision to join CAM Systems. 67% indicated that when they consider job offers, if the benefits package is not competitive or better than competitive, they are not interested. 17% went as far as suggesting that benefits are their most important job selection criteria.

	"Think back to the days when you were considering various job offers. To what extent did this benefits package influence you to join CAM Systems?"

Employees gave favorable ratings on the amount, quality and accessibility of the information they were receiving on benefits, and on the degree to which both medical and non-medical benefits were meeting or exceeding their expectations. They also gave favorable ratings to both outcome variables. The rating point of 5 indicates "somewhat agree" on the seven point agreement scale.

Pearson correlation matrices were computed to explore relationships between the outcome variables and the information and expectations categories. As expected, the expectations categories had statistically significant, positive relationships with benefits information, reputation and employee appreciation (all correlations were statistically significant, between .40 and .70). In short, employees who felt their expectations were being met also tended to give CAM Systems a superior reputation for competitive benefits and felt that CAM Systems values and appreciates its people. Similarly, employees who felt well informed on their benefits packages tended to feel more valued and appreciated as well.

What was not expected were statistically significant, positive relationships between benefits information and the expectation variables (correlation with medical benefits expectations .49; with non medical benefits expectations .44), as illustrated below:

This result is surprising, given these benefits packages are not perfect, nor were they designed to be. Perfect benefits packages are simply unaffordable, making the benefits strategy one of satisficing, not optimizing. CAM Systems wants to offer its employees superior value, recognizing that some features can not be as generous as all their employees might prefer. These areas were criticized in the written comments sections of the survey, which solicited both positive and negative feedback on each benefits package. Every major benefits package had easily recognizable inherent constraints and limitations, without exception:

Health Plans & Services. 582 complaints focused on various aspects of medical, dental and visual benefits, consistently targeting high co-pays, limitations in coverage (treatments and choice of providers) and poor customer service (responsiveness, claim rejections, and paper work problems)

Income Protection. 255 comments complained about the features of various forms of insurance coverage, retirement programs and reimbursement programs for health and dependent care. 401(k) programs and online financial advice were preferred targets (141), due to choice restrictions and lackluster stock market performance.

Work Life Benefits. 303 complaints assessed various work life benefits, ranging from legal plans and domestic partner benefits to service awards. Health and fitness benefits were particularly vulnerable (135), because accessibility varies widely across corporate locations.

Time Off Benefits. 345 complaints focused on issues around leave time (sick, personal, maternity, paternity), holidays and vacations. Vacation policies of 2 weeks off for newer hires were a preferred target, since they are perceived as somewhat below industry average (228).

The assumption that well-informed employees would understand these weaknesses better than less informed employees, and lower their ratings accordingly, was not accurate. Informing employees of the details of the benefits programs, limitations and all, improves the favorable ratings these benefits subsequently receive.

These findings indicate that in the absence of accurate information on CAM Systems benefits, other information sources - impressions of industry norms, previous experience at other companies, vicarious experience gleaned from informal conversations, etc. - depress expectation and outcome ratings. In short, employees who feel well informed and understand their benefits packages tend to rate CAM systems benefits as meeting or exceeding their expectations, and as being more competitive with the benefits of other firms.

Bottom line: Tell all the employees, all the time, the truth about benefits
warts and all; the return is rewarding.

Contact PDi about these publications

(copyright 2002 by PDi. All rights reserved. Please do not modify, copy or distribute this article -- see website terms and conditions of use.)