Chapter 5 stressed the crucial role of managers in shaping whistleblowing outcomes through their important influences on whether or not public sector employees would ultimately view reporting in a positive or negative light. The previous section found two key risk factors associated with managerial responses that come to be perceived as mistreatment, increasing the likelihood by a magnitude of four that a whistleblower will perceive their treatment this way: first, if for whatever reason the reporting path has gone external, and second, if the event has failed to result in a positive substantive outcome. This second risk factor was also prominent among those associated with co-worker mistreatment, where its presence made the odds of poor treatment more than three times greater.
The final analyses in this section use these high-risk indicators to identify any factors associated with employee survey cases at the extreme ends of the risk spectrum. First, an analysis can be undertaken to examine the cases of ‘higher-risk’ respondents for whom both these high risk factors were present, but who, against the odds, did not record mistreatment by co-workers and/or management. By comparing these ‘higher-risk, good-outcome’ respondents with higher-risk respondents who recorded the more predictable bad outcomes, it is possible to identify any reasons why even in difficult cases some reporters manage to emerge relatively unscathed.
Second, the reverse analysis can be undertaken to examine ‘lower-risk’ respondents who reported wrongdoing, for whom neither of the high risk factors were present and whose cases should therefore have been much easier for agencies to manage to a positive conclusion, yet who nevertheless still recorded mistreatment. By comparing these ‘lower-risk, bad-outcome’ respondents with lower-risk respondents who did not record bad outcomes, it is possible to identify some reasons why even more straightforward cases still result in problems. Both analyses provide insights into how whistleblowing cases might be better managed.
Because the groups in each case are relatively small, these analyses are not limited to public interest whistleblowers; rather, they take in all non-role reporters, including those who reported personnel and workplace grievances. The extent to which this can influence the outcome is discussed. As the statistical tests for this section could not be repeated using logistic regression, alternative significance tests were used.
When employees who report wrongdoing are at higher risk of mistreatment but, against the odds, do not record it, are there distinctive factors that explain these better-than-expected outcomes? Such cases are clearly valuable for their potential insights into how whistleblowers might be more effectively protected in a wide range of high-risk situations. In these cases, the report did not result in any substantive improvement in relation to the alleged wrongdoing, despite the reporter having pursued the matter outside the organisation, yet the reporter recorded being treated well or the same by co-workers and management as a result of the process.
Table 6.8 sets out the features of this group, comparing first those who recorded good and bad treatment by management, and secondly those who recorded good and bad treatment by co-workers. The group is small (n = 105), making up less than 7 per cent of all reporters. Nevertheless, these respondents were spread widely across the employee survey agencies, suggesting this less common combination of outcomes could potentially arise in any agency. Despite the fact that the analysis now includes all reporters, including those whose reports concerned personnel or workplace grievances, all the risk factors identified in the previous section hold, but now in reverse. For those treated well, it remained the case that the wrongdoing was perceived as less serious, occurred less frequently, was less likely to have been directed at the reporter personally, was less likely to have involved multiple wrongdoers or wrongdoers more senior than the reporter and involved an immediate work unit of 20 or more people.
Only one new factor emerged as providing additional ‘protection’ for reporters, further reducing the likelihood of mistreatment by co-workers, notwithstanding the otherwise high risks of this occurring. This factor was the reporter’s age, with reporters who escaped mistreatment by co-workers being significantly older (mean of 46 years) than those who recorded mistreatment (mean of 40 years). This result suggests that, again, power differentials are in play, with older employees better able to ride out the storm even when the risks of mistreatment are otherwise high. In all other respects, however, no magic answer emerged from this analysis. This result confirms that, if agencies are to minimise the risks of whistleblower mistreatment, this can reliably be pursued only through policies and processes that offset the many situational and outcome variables that currently contribute to poor outcomes and by actively intervening in the normal operations of the organisation, rather than letting ‘natural’ organisational processes take hold.
|
Means and frequencies |
||||
|
Risk factor |
(1a) Treated well/same by management (n = 33) |
(1b) Treated badly by management (n = 71) |
(2a) Treated well/same by co-workers (n = 81) |
(2b) Treated badly by co-workers (n = 24) |
|
The wrongdoing was frequent 1 = just this once 5 = all the time |
3.24** |
3.82** |
3.52* |
4.04* |
|
The age of the wrongdoer (years) |
n.s. |
n.s. |
46.21* |
40.23* |
|
How became aware of wrongdoing? |
||||
|
Directed at me (n = 60) |
15% (9)*** |
85% (51)*** |
n.s. |
n.s. |
|
Not directed at me (n = 44) |
55% (24)*** |
45% (20)*** |
n.s. |
n.s. |
|
How many people involved? |
||||
|
One (n = 19) |
n.s. |
n.s. |
95% (18)* |
5% (1)* |
|
A few/large number/ widespread (n = 85) |
n.s. |
n.s. |
73% (62)* |
27% (23)* |
|
Position of wrongdoer |
||||
|
Below or at same level (n = 23) |
57% (13)** |
44% (10)** |
n.s. |
n.s. |
|
Higher level (n = 81) |
25% (20)** |
75% (61)** |
n.s. |
n.s. |
|
Size of immediate work group |
||||
|
Less than 20 (n = 57) |
21% (12)* |
79% (45)* |
n.s. |
n.s. |
|
20 or more (n = 46) |
43% (20)* |
57% (26)* |
n.s. |
n.s. |
n.s. = not significant
* difference statistically significant at p < 0.05
** difference statistically significant at p < 0.01
*** difference statistically significant at p < 0.0005
Source: Employee survey: Q21, Q23, Q24, Q25, Q39, Q49.
When employees who report wrongdoing are at lower risk of mistreatment but nevertheless still record it, are there distinctive factors that continue to explain these worst of all outcomes? Such cases might be considered a classic ‘reprisal scenario’, in which the reporter loyally kept the matter internal to the organisation and regarded their report as having been vindicated, but they were still made to suffer. In this case, the perception of mistreatment cannot be attributed simply to whistleblower dissatisfaction with the outcome, because there is less basis for conflict about the truth of the matter or frustration over inaction. What factors are associated with these outcomes?
Table 6.9 sets out the features of this group, comparing first those who recorded good and bad treatment by management and, second, those who recorded good and bad treatment by co-workers. Despite the fact that the analysis now includes all reporters, all the risk factors identified in the previous section still hold. For those treated poorly, it remained the case that the wrongdoing was perceived as more serious, occurred more frequently, was more likely to have been directed at the reporter personally and involved wrongdoers more senior than the reporter, with more than one wrongdoer involved.
Three previously unidentified factors also appear to be significant in the poor outcomes of these lower-risk reporters. As none of these appears unique to personnel or workplace grievances, they can also be used to interpret the results for public interest wrongdoing.
First, even though all reporting remained internal for these groups, reporters were more likely to record mistreatment by management and co-workers if their report went through a larger number of internal stages before leading to positive action. When the number of internal reporting stages was restricted to one, reporters were much less likely to indicate poor treatment. This supports Miceli and Near’s (1988) finding that allegations of wrongdoing that were resolved after the first report were less likely to be associated with reprisals, but that whistleblowers who were required to make their allegations in successive reports (usually to higher levels of authority) were more likely to experience retaliation. These findings further reinforce the importance of a competent initial response to a disclosure, preventing the need for a second or further report, if the chances of protecting even lower-risk whistleblowers are to be maximised. Given the evidence in Chapter 4 that most employees’ first reaction is to report wrongdoing to their supervisor or another manager, irrespective of whether this is always wise, the importance of the first investigative response also reinforces the need for agencies to increase efforts for ensuring that disclosures reach the right place.
|
Mean and frequencies |
||||
|
Factor |
(1a) Treated well/same by management (n = 390) |
(1b) Treated badly by management (n = 20) |
(2a) Treated well/same by co-workers (n = 390) |
(2b) Treated badly by co-workers (n = 20) |
|
The wrongdoing was serious 1 = not at all, 5 = extremely |
3.89** |
4.45** |
3.88**** |
4.63**** |
|
The wrongdoing was frequent 1 = just this once, 5 = all the time |
2.64** |
3.30** |
n.s. |
n.s. |
|
How became aware of wrongdoing? |
||||
|
Directed at me (n = 104) |
89% (92)**** |
12% (12)**** |
n.s. |
n.s. |
|
Not directed at me (n = 306) |
97% (298)**** |
3% (8)**** |
n.s. |
n.s. |
|
How many people involved? |
||||
|
One (n = 215) |
n.s. |
n.s. |
98% (211)** |
2% (4)** |
|
A few/large number/ widespread (n = 183) |
n.s. |
n.s. |
92% (168)** |
8% (15)** |
|
Position of wrongdoer |
||||
|
Below or at same level (n = 253) |
98% (248)**** |
2% (5)**** |
98% (244)*** |
2% (5)*** |
|
Higher level (n = 149) |
90% (134)**** |
10% (15)**** |
90% (132)*** |
10% (14)*** |
|
Number of reporting stages |
||||
|
Only 1 (n = 236) |
98% (231)** |
2% (5)** |
97% (222)* |
3% (6)* |
|
More than 1 (n = 174) |
91% (159)** |
9% (15)** |
93% (161)* |
8% (13)* |
|
Investigation into report |
||||
|
Yes (n = 385) |
96% (369)** |
4% (16)** |
n.s. |
n.s. |
|
No/don’t know (n = 21) |
81% (17)** |
19% (4)** |
n.s. |
n.s. |
|
Employment status |
||||
|
Full-time (n = 358) |
96% (344)* |
4% (14)* |
n.s. |
n.s. |
|
Part-time/contractor/ casual (n = 52) |
89% (46)* |
12% (6)* |
n.s. |
n.s. |
n.s. = not significant
* difference statistically significant at p < 0.05
** difference statistically significant at p < 0.01
*** difference statistically significant at p = 0.001
**** difference statistically significant at p < 0.0005
Source: Employee survey: Q21–5, Q28, Q33, Q46.
Second, the likelihood of lower-risk reporters recording management mistreatment was increased when—even though the problem was addressed—this was either without a formal investigation or without any of which the reporter was aware. When the reporter was confident that an investigation was held or, in other words, that it was their disclosure that led to the matter being addressed, close to only one in 20 from this group recorded managerial mistreatment. The mistreatment rate increased, however, to one in five of the group if the disclosure was either not investigated or the response was unknown. Given that the entire group believed that overall things became better in the organisation, it is likely that there was an investigation or other action, but clearly these reporters were not kept informed of its progress or outcome.
This result is consistent with the evidence in Chapter 5 that a failure to keep whistleblowers informed of the organisational response can contribute directly to stress and other negative outcomes. It is also likely to be an indicator of an environment conducive to direct reprisals, given that the whistleblower has not been armed with the information necessary to claim vindication and, indeed, management might have addressed the problem without any recognition or support of the whistleblower whatsoever. This result can therefore indicate a situation in which the whistleblower is correct but is either ‘left in limbo’ or ‘hung out to dry’.
The third additional factor is the employment basis of the reporter. Lower-risk reporters who were nevertheless still treated badly by management were more than three times more likely to be employed on a part-time, casual or contract basis. Even when the level of conflict surrounding a report is otherwise low, the reporter’s relative position in the organisational structure continues to be a major risk factor for reprisal. This result is consistent with other more general findings, here and elsewhere (Miceli et al. 1999), that the power of a whistleblower relative to that of wrongdoers is a direct determinant of the likelihood of negative treatment. When considering which cases deserve additional support and protection, it is imperative that agencies focus on low-status employees even when the risks otherwise appear low.
This section has highlighted that, for those whistleblowers with extreme levels of risk (either high or low), a number of additional risk factors come into play. If an investigation into a disclosure is kept internal to the organisation and things become better as a result, this ordinarily suggests a good outcome for the whistleblower; this group can therefore be considered ‘low-risk’ reporters. These analyses show, however, that, even if an investigation is conducted internally and is resolved, it is important that this occurs as early as possible (so that the whistleblower does not have to report more than once, if at all possible) and for the whistleblower’s role to be recognised through advice and information. For high-risk whistleblowers (those with an unsuccessful outcome, in which the investigation was external to the organisation at some point), the results reinforce that the key to obtaining reasonable outcomes lies in assessing and compensating for the power differentials between reporters and suspected wrongdoers.