Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Change format of listening survey responses #116

Open
JamesPHoughton opened this issue Feb 14, 2023 · 7 comments
Open

Change format of listening survey responses #116

JamesPHoughton opened this issue Feb 14, 2023 · 7 comments

Comments

@JamesPHoughton
Copy link
Contributor

JamesPHoughton commented Feb 14, 2023

In pilot tests with the existing Listening surveys (listeningQualityOwn, and listeningQualityPartner) we use a likert scale from "Not at all" to "Very Much" for each question:

image

These measures seem to saturate, possibly because:

  • "very much" is the desirable answer, in terms of what people ought to do, and society expects them to be able to do
  • "very much" may not capture the full range of outcomes - people aren't perfect at these skills even if they do try very much

By changing how the response is collected, we may be able to change where people anchor on the scale so that we actually get some variation rather than saturating the measure. One way to do this is by increasing the meaning of the highest value so that it is an unreasonable expectation. Nobody is likely to claim that they listen perfectly 100% of the time.

image

We can also use a midpoint indicator to intentionally anchor people's expectations about what is normal a little lower in the space:

image

I'd also like to make this survey more general so that it could be used for discussions with more than two people, hence moving towards "others" rather than "my conversation partner"

@JamesPHoughton JamesPHoughton changed the title Change format of listening surveys Change format of listening survey responses Feb 14, 2023
@JamesPHoughton
Copy link
Contributor Author

@NettaWeinstein - would love your thoughts on these suggested changes!

@JamesPHoughton
Copy link
Contributor Author

JamesPHoughton commented Feb 14, 2023

Possible full survey:
screenshot

For some reason the right hand side is cut off..

@JamesPHoughton
Copy link
Contributor Author

This is also super long - do we have data from previous iterations that might let us collapse some of the question items?

@JamesPHoughton
Copy link
Contributor Author

JamesPHoughton commented Feb 15, 2023

Going back to the original source: Kluger, Avraham N., and Osnat Bouskila-Yam. 2017. “Facilitating Listening Scale (FLS).” In The Sourcebook of Listening Research, 272–80. Hoboken, NJ, USA: John Wiley & Sons, Inc. reveals that these are a subset of a larger listening behaviors survey, which is fully:

For each item, we want you to reflect on your experience of being listened to by your supervisor. For each item, indicate your level of agreement/disagreement using the following scale:

  • 1 = Strongly Disagree
  • 2 = Disagree
  • 3 = Somewhat Disagree
  • 4 = Neither Agree nor Disagree
  • 5 = Somewhat Agree
  • 6 = Agree
  • 7 = Strongly Agree

The following items measure perceptions of listening behaviors and should be introduced by the prompt, “When my current supervisor listens to me, most of the time, s/he...”

Constructive listening behaviors

  1. Tries hard to understand what I am saying
  2. Asks questions that show his/her understanding of my opinions
  3. Encourages me to clarify a problem
  4. Expresses interest in my stories
  5. Listens to me attentively
  6. Pays close attention to what I say
  7. Gives me time and space to talk
  8. Gives me his/her undivided attention
  9. Creates a positive atmosphere for me to talk 10) Allows me to express myself fully

Destructive listening behaviors

  1. Talks offensively
  2. Criticizes my feelings
  3. Frowns (showing disapproving facial expressions)
  4. Discounts or explains away my feelings
  5. Is not willing to listen to me
  6. Does not pay attention to things I say
  7. Talks back to me aggressively
  8. Becomes irritated
  9. Is impatient
  10. Gets tense

Constructive listening behaviors – reframing

  1. Restates what I say
  2. Gives me a brief summary of what I have said
  3. Completes my sentences to help me clarify what I am saying
  4. Asks continuing questions like “Could you tell me more?”

Destructive listening behaviors – domineering

  1. Often interrupts me while I am talking
  2. Begins to talk before I finish talking
  3. Talks more than me
  4. Imposes his/her own views
  5. Listens to me calmly (reverse scored)
  6. Hurries me into talking faster

Destructive listening behaviors – escaping

  1. Stares at the computer screen while I’m talking to him/her
  2. Uses the telephone while I’m talking to him/her
  3. Is distracted while I’m talking

Destructive listening behaviors – no time

  1. Begins a discussion by telling me how long s/he has for me
  2. Looks at his/her watch or clocks in the room when s/he has limited time to listen to me
  3. Hurries me and lets me know that s/he has a limited amount of time to listen

Destructive listening behaviors – changing the subject

  1. Starts talking about unrelated issues
  2. Changes the subject too frequently

What we have so far is just the constructive listening items. We may be interested in including some of the other items with the appropriate coding?

Also, the original paper talks about a study Itzchakov, G. (2015, May). If you listen to me, I will change my attitude. In A. N. Kluger (Chair), Listening: Why should you and why should you not? Symposium presented at the 30th Annual Conference of the Society for Industrial and Organizational Psychology, Philadelphia, PA. that just used "The seven items measuring constructive listening behavior that loaded most highly in Bouskila‐Yam and Kluge" - its not clear what these items were though, although I think we know the Guy to ask? If we can get the factor loadings from that original study, we might be able to use those to condense the survey.

@JamesPHoughton
Copy link
Contributor Author

@NettaWeinstein, thinking more about the midpoint label after our discussion today, I think you're right to make a distinction about "About Average" as something that makes the reader form a comparison with what they would have otherwise seen in their daily lives - and it could be all over the place in terms of what people expect, as people have different exposure and experience of other people's listening skills. Adding a sentence to say "Taking into account that this is your first discussion with this group..." tries to narrow down the frame of reference people are using to make comparisons against, to reduce noise that comes from differences of experience or interpretation of what they should be comparing to.

The framing of "what fraction of the time..." has a bit more of an objective feel to it, and would be great for questions like "listens to me attentively", but maybe less so for behaviors that are intermittent, like "Asks questions that show their understanding...".

The framing of "Adequately" or "An acceptable amount" changes the frame of reference to what the participant thinks their interlocutor should have done, with the left-hand side of the scale implying they should have done it more, and the right-hand side implies going above expectations.

We could also go back to the "strongly agree... strongly disagree" framing from the original.

@JamesPHoughton
Copy link
Contributor Author

I've got a bug in the slider implementation at the moment (If the user doesn't enter a value, it submits the midpoint (!!) without telling them...).

For this week's tests, lets use the original likert scale (Strongly Disagree... Strongly Agree) frame and see what we get.

@NettaWeinstein
Copy link

NettaWeinstein commented Feb 16, 2023 via email

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants