Is it a repeated measurements multilevel modeling?

Welcome to the forum for MLwiN users. Feel free to post your question about MLwiN software here. The Centre for Multilevel Modelling take no responsibility for the accuracy of these posts, we are unable to monitor them closely. Do go ahead and post your question and thank you in advance if you find the time to post any answers!

Remember to check out our extensive software FAQs which may answer your question: http://www.bristol.ac.uk/cmm/software/s ... port-faqs/
Post Reply
nasirihakimeh
Posts: 4
Joined: Mon Jul 26, 2021 12:07 pm

Is it a repeated measurements multilevel modeling?

Post by nasirihakimeh »

Hello everyone,
Please see the attached model and screenshots of a sample of the collected data as I was not allowed to attach an excel file. I would highly appreciate your time and consideration if you would share your thoughts about our study.

Please read the details here:
It is conceptualized in the tourism field that there are 6 different and distinct aesthetic experiences:
1) experience of the beautiful (i.e. visiting a place which makes you feel that it is beautiful,
2) experience of the ugly, and 4 more different experiences in between the beautiful and the ugly.

We want to know:
1: What is the effect of aesthetic features of a place (e.g. crowdedness, harmony, …) on aesthetic experiences (i.e. experience of the beautiful, the ugly, and so on)
2. What is differences and similarities in relation to how destination aesthetic features are experienced by tourists and residents in the occurrence of 6 different types of aesthetic experiences

The variables of our interest are:
Dependent variables:
1. The frequency of occurrence of the experiences (7 points Likert scale).
Note: For measuring this variable, the equivalent question in the survey reads like:
How often have you seen a place/ thing that made you feel like the experience of the beautiful?
 Never
 Very Rarely
 Rarely
 Occasionally
 Often
 Frequently
 Always

Independent variables:
1. The individual's evaluation of perceived Destination’s Multidimensional Aesthetic Features (i.e. an 18 items scale with 6 dimensions)
2. Belongingness to a group (tourist vs resident)

Control variables:
1. Age
2. Gender
3. Education
4. Residents' district of living/ Tourists' city of residence

Moderator:
Aesthetic centrality: (i.e. an 11 items scale with 3 dimensions)

We've used a cross-sectional survey and only one time (not under different numbers of measurement occasions) we asked respondents to evaluated the features of one of the places (on a 7 point Likert scale) that they felt like the experience of the beautiful. We asked the same questions about the other 5 types of aesthetic experiences.
All the places (i.e. tourism attractions) are located in a particular city. So 6 different experiences happened in so many different tourism attractions. That is, it is not 6 distinct tourism attractions as the place where the experiences have been actualized.



With warm regards,
Hakimeh
Attachments
20210724_multi-level model.jpg
20210724_multi-level model.jpg (130.45 KiB) Viewed 741 times
Data.xlsx
(16.61 KiB) Downloaded 32 times
billb
Posts: 146
Joined: Fri May 21, 2010 1:21 pm

Re: Is it a repeated measurements multilevel modeling?

Post by billb »

HI Nasirihakimeh,
Apologies that I don't fully follow your structure. Do you have people's views nested within places or places nested within people's views? I think the main thing is with only 6 places you are probably looking at fitting fixed effects for these as they are probably important rather than just a random set of places?
Maybe if you can be clearer on your design?
Best wishes,
Bill.
nasirihakimeh
Posts: 4
Joined: Mon Jul 26, 2021 12:07 pm

Re: Is it a repeated measurements multilevel modeling?

Post by nasirihakimeh »

Hi Bill,
I appreciate your response. Thank you for giving me the opportunity to clarify my question further. Please let me know if I need to provide more clarity.
Actually, we don't have either people's views nested within places or places nested within people's views. We have experiences nested within people.

To clarify more, may I sincerely ask you to see the following. It is one section of our survey.
****************************************************************************************************************************
****************************************************************************************************************************

Part 1
In this section, we will present you with descriptions of six types of experiences that you may have encountered while exploring Kish (i.e. the name of the city). We are interested in the frequency of occurrence of each experience and the features of one of the places that made you experience as such.

Read the following description and try to remember one place in Kish that you had such an experience.
1. Experience No. 1
“Visiting an ugly place, you feel disgust, shocked, confused and baffled. At that moment, your mind and body become uncomfortable. You feel overwhelmed with anxiety, disappointment, and annoyance. You feel disrespected and disconnected with the place and the people. At the same time, you feel pity for other people who have to live in such a place, and you have a desire to leave that place.”

First: During recent days that you have been traveling in Kish, how often have you seen a place that made you feel like the experience No. 1?
 Never
 Very Rarely
 Rarely
 Occasionally
 Often
 Frequently
 Always

Second: Write the name of one of the places that you had such an experience. ……………………………………

Third: Rate the features of the place that you mentioned above on a scale of 1 to 7.

1. I would say that the place was: not crowded 1 2 3 4 5 6 7 crowded
2. I would say that the place had: nothing to see 1 2 3 4 5 6 7 lots of things to see
3.
.
.
.
18. I would say that the place was: ordinary 1 2 3 4 5 6 7 unique
****************************************************************************************************************************
****************************************************************************************************************************

Please note that:
1) The 18 above-mentioned questions correspond to 18 items scale with 6 dimensions that you can see in the previously attached model as "Destination Aesthetic Qualities"
2) In the above section, experience No. 1 refers to the experience of the ugly. The description comes from another study of us adopting descriptive phenomenology to describe the experience of the ugly, the beautiful, and other 4 types of aesthetic experiences in between the beautiful and the ugly.
3) The same set of questions will be repeated 5 times asking people's views about the features of one of the places in the occurrence of another experience (i.e. the beautiful, the sublime, the picturesque, and so on).
4) The places that the experiences have been actualized are random.
billb
Posts: 146
Joined: Fri May 21, 2010 1:21 pm

Re: Is it a repeated measurements multilevel modeling?

Post by billb »

Hi Nasirhakimeh,
This is quite unusual data and your question is only a little about multilevel modelling so I'll give a brief answer of what I might do and then leave to see if others have ideas.
So perhaps a straightforward thing to do is to consider each of the 6 types separately e.g. here are a series of individual responses to an 'ugly' place in Kish. You would then have a 2 level structure with individuals nested within 'chosen' places and your predictor variables could be your 18 questions, whether respondent is tourist/resident and their overall impression of Kish (Q1). The main challenge I can see is I am not a big fan of your questionnaire as it asks them to pick ONE place so I suspect you'll have large biases towards the places that people find most ugly rather than ugly places per ce but I guess that is what you are left with analysing.
I guess in theory you could try and then combine the 6 responses but I don't think you'll gain much aside from making life difficult for yourself as the dependency structure gets complicated as each response is (possibly) a different place so would be a cross-classification of place and person and presumably you will put lots of interactions as you would want different effects for predictors on each response. I guess if you can figure precisely how you want to do this you might be able to add some correlations but would depend on your research question.
Best wishes and good luck,
Bill.
nasirihakimeh
Posts: 4
Joined: Mon Jul 26, 2021 12:07 pm

Re: Is it a repeated measurements multilevel modeling?

Post by nasirihakimeh »

Hi Bill
Thanks a lot for your insightful comment.

The issue is that there are no particular 'chosen' places. Respondents have named so many random places in that particular destination. What makes it more critical is that sometimes there are overlaps among some places as the setting of different types of aesthetic experiences. For example, one respondent experienced the beauty in place A, and the other one experienced the picturesque in the same place. So let's forget about the places.

Is there any way to think of it as "the effect of aesthetic features of a place on the frequency of occurrence of experience of the ugly " is nested within individuals? And the same for the other 5 experiences? Does it make sense?
If not, do you think that my data structure is not appropriate for multilevel modeling?

******************************************************************************************************
FYI, please!
Research objectives:
1. To associate destination aesthetic qualities with the occurrence of individuals’ aesthetic experiences (i.e. six distinct aesthetic experiences)
2. To understand differences and similarities in relation to how destination aesthetics are experienced (the beautiful, the sublime, the picturesque, the ugbeaful, the beaugly, and the ugly) by tourists and residents.


Many thanks,
Hakimeh
billb
Posts: 146
Joined: Fri May 21, 2010 1:21 pm

Re: Is it a repeated measurements multilevel modeling?

Post by billb »

Hi Hakimeh,
The challenge you have is so much missing data. For any 1 place lets say that 2 people say it is beautiful and 1 picturesque out of 100 people. What you don't know is the opinions of the other 97 as you only ask them to choose 'an example' so they may all think the place in question is beautiful for example including the 1 who chose 'picturesque' but just chose a different example and also people might decide they need to choose different places each time (human nature) so you are less likely to het the same person choosing the same place twice.
You only have 1 place of a type nested within individuals so that is not multilevel - the only nesting I can see is when you have repeats for the same place.
Best wishes,
Bill.
nasirihakimeh
Posts: 4
Joined: Mon Jul 26, 2021 12:07 pm

Re: Is it a repeated measurements multilevel modeling?

Post by nasirihakimeh »

Hi Bill,

I really appreciate your time and consideration. I got it, the structure of our data is not appropriate to adopt repeated measurements multi-level modeling.

As an alternative way to analyze data, as we have already collected almost 300 samples, six regression models can be used to explore the effects of destination aesthetic features on six different types of tourism aesthetic experiences. For calculating each model, the frequency of occurrence of each aesthetic experience will be used as the dependent variable. The candidate independent variables will be six dimensions of destination aesthetic features.

May I ask your opinion about the above alternative, please?

Warm regards,
Hakimeh
Post Reply