Phone Polling
 

Why are we doing phone polling only? An online poll can be completed at any time that is convenient for the people being polled and an online poll will not be blocked by my phone. I just saw the email from the union about it showing up as “Spam Likely.” I have received numerous of those calls and ignore them. How much are we paying to have people call and ask how likely we would support something? It’s time to move to the 20th century and do online polling.

I'm guessing they'll say self selection bias. That said, I'm sure they could easily accomplish the same thing with an online survey emailed to the sampling groups, along with the phone calls. I don't mind them calling, they just have to have the ability to call back at a time that is better for me (which I've heard they will do). Just put the number in your contacts and it won't show up as spam.

Exactly. Unless you get a very high percentage of respondents (over 60%) the data is pretty skewed. (Think CNN or FOX online poll)

That being said, there needs to be a way to hit the people like me who send almost all calls to VM that aren’t known..

The phone number is in the e-mail. Make it a contact and it will be known.

The issue of not enough participants and skewing data isn’t unique to online polling. That is true with any poll that only samples a small portion of a group.

True, but scientific polling can take a much smaller sample with a much more accurate result.

I think this might be true for narrowly tailored surveys.

Think "Who will you vote for?" or "What is the thing you like/hate most about our brand."

In my experience, ALPA tries hard, but the polling is usually all over the place with leading questions and insane hypotheticals. There's just too many variables when you're dealing with 14,000+ pilots who all want different things.

Better than nothing, I guess.

I would agree. Only so many questions you can ask on a poll, or so many details you can get into.

The Dalpa pollster is outstanding and has very unbiased questions. The problem arises when the Reps start don’t get the result they expect and start forcing questions on him.

This was very evident in last years polling.

What makes phone polling vs online polls scientific?

They did an online survey. Not enough people took the time to fill it out. They had to extend the deadline multiple times trying to get higher participation.

Phone polling works much better with a small sample size. It’s very accurate.

So participation rates are higher with phone? I know the online poll for scheduling was around 60% if I recall.

I was wondering who was calling me everyday. My spam blocker app kept blocking it.

Participation rates for phone polling are much lower. It’s a controlled sample so you don’t need to have a a large number as self selection bias is minimized.

Can you explain controlled sample? Are they targeting specific groups for phone polling and not just doing random samples?

A controlled sample inherently means random. In order to get any useful data, they have to sample pilots from across the company.

Within the context of ALPA's polling? It's done by a very experienced pollster who ensures the sampling is done by demographic. Age, category, Regular, Reserve, commuter, etc. The larger the sample, the more accurate the data. Another control is the specificity of the question(s). The more binary the question (yes or no), the more accuracy.

The way is was explained to me, the pollster takes known demographic information, then develops a "call order" for which pilot to contact first, second, third, etc. within each group. Then the goal is the get as much feedback quickly before other events change the perspective of respondents. "How would you rate your contract compared to peer airlines?" changes the day UAL, American, or others get a deal, or the stock market crashes...so they try to get "same day" (my term) responses. So the timing is the final element of "controlled sample".

Based on the comments some reps have made in the past regarding polling, there are some who don't trust data gathered scientifically. I asked my former Capt rep in MSP what his problem was with the polling, and his reply was, "It's too neutral!"

Thanks. I guess I didn’t realize it was targeted to demographics.

I was called on the last round each time I was busy and unable to take the call. But I did fill out the online survey, personally I'd much rather do that because then I'm not in a time constraint and can answer when I have time.

Contrasting that with an online survey, the more passionate you are, the more likely you are to respond. So a so called “vocal minority” can skew the results. The problem is that while they are vocal, if you don’t have a huge response rate (>75%) you don’t know if they are a minority or not. You can correct for this somewhat by doing phone polling at the same time but it’s not perfect. Online surveys are great for getting ideas and long-form feedback but that’s about it.

In my opinion the people we are paying insane money for phone polling have sold us that free online polling is no good.

Let’s make it simple.

Poll every Delta pilot online. If the pilots who don’t bother get the short straw, they need to look in the mirror.

Phone polling is so 1960s.

I guess in my naïveté I assumed we did that.

Methods and statistical confidence.

I would think the best process would be a combination of phone and online polling. For example, the selection process is still done by phone. If you don’t answer they could leave a callback number to take the poll at convenience or perhaps provide a unique PIN code and link to go answer the questions online. This way, the selection process continues to be a “random sample” rather than a regular open poll that may not be answered by a cross section of the group.


Sent from my iPhone using Tapatalk Pro

Theoretical sample and actual sample are two very different things. My wife was a marketing research analyst and she used to hate the inherent data collection problems with phone sampling. 9-5 M-F to lines where people actually answer after caller ID was 90% retirees. After business hours even fewer answered because of the telemarketing hours of 5-8. Weekends, fugedabowit. The data centers end up adjusting the calling parameters just to get a large enough sample and fill the responses by any means to meet a deadline.

Well for people like us who interact with a message forum like this, a online survey seems easy. But for the older crowd, or those who don't like technology as much, not as many would fill it out. Then, when the MEC tries to use the data to establish priorities, some LEC members might complain that certain demographics weren't represented. I think saying "tough" doesn't fly.

When I talk to Captains I fly with about the surveys, and they say they don't answer the phone, my response is, "fine, I do and that just means my opinion is considered and yours isn't". It's their choice to answer or not.

But that still doesn't provide valuable data. Just because a pilot doesn't take a survey doesn't mean they won't vote. If you ignore their opinion because they didn't remember or have motivation or couldn't remember their log in, etc to do a usually monstrous survey online does not mean they will just have to accept the contract they get. They will vote and if you just "give them the short straw" they will vote no. And all that time and money was completely wasted.

It's far more important to get an actual feeling of the pilot group over the ones that are most outspoken. They get just as many votes as those of us who have more important things to do than fill up forums with retirement restoration nonsense.

Sent from my SM-G975U using Tapatalk

Vocal minority? Along with R.O.U.S’s, I don’t think they exist. :)

Scoop

Careful, Scoop. you know what happened to The Dread Pirate Roberts right after he said that... :p

Inconceivable!

I do not think that means what you think it means.

If they can choose to call a certain phone number to reach an individual of a specific demographic, surely they could selectively email an online poll to specific individuals just as easily. They also could include data from only the first X respondents of a given demographic if they want that control over equally representing the different demographics.

The last part doesn't solve for self-selection bias.

Sent from my SM-G975U using Tapatalk