STC Salary data collection

[Ned Bedinger]

Steven Jong wrote:

> 
> The article in the November 2007 issue of Intercom describing the BLS  
> data gives a number of reasons for the discrepancy, and I'd like to  
> touch on one reason in particular. Studies have shown that employees  
> tend to inflate salaries when responding to salary surveys. 

It would be fun to design a survey that weighted responses according to 
how well the reported salary was documented.  A redacted copy of a pay 
stub would have less weight than an unredacted one, etc.

It seems like it would be easy to maintain anonymity AND get some 
percentage of validated data by asking respondents to take a tax or pay 
stub, along with STC card and ID to a notary, and have the survey form 
'validated.' As long as the survey had a gold standard like this, I 
think you could get by with a small percentage of notarized responses.


> Some  
> people in other forums have bristled at that suggestion, but I don't  
> think we should take it personally. First, the statement is a general  
> observation, not directed at technical communicators and certainly not  
> at STC members. 

Well, it kinda is directed at STC members, since iirc it has a lot to do 
with the reason for changing things.

> Second, the STC surveys were self-selecting  
> (respondents chose to respond), and self-selection explains how  
> inflation can happen even if every respondent answers honestly. 

All voluntary surveys are self-selecting.  Still, I'm not sure that 
professional tech writers tend to treat their participation in their own 
salary survey as a referendum on what their salary ought to be. But 
perhaps I overestimate my colleagues. Maybe we're very politically 
motivated (politics of pay), but not especially inspired, except by the 
pay packet??


> If you  
> feel good about your salary, and have reason to believe it's above  
> average among your peers, you are more likely to report it;  
> conversely, if you feel you are underpaid, you are less likely to  
> report it. 

My logic says that the survey is a communication between the tech writer 
and the mothership, not with employers, so respondents would not be 
inspired about faking up ammo to use on HR, but rather would be looking 
forward to a constructive dialog about pay with their peers, mentors, 
experts, and more-seasoned veterans.  I'm curious (non-member): is there 
an effort made to do this with members, especially noobs, after the survey?

> Self-selecting surveys are known to be biased by survey  
> professionals; 

No, professionals worthy of the name would not do that.





:-)

Ned Bedinger
doc at edwordsmith.com