How did you word your hypotheses for these two constructs? Do employee feedback and the local near have a significant and negative influence on recomm? Are the measurement items formulated in the negative form?

If the answer to these questions is no and the orientation of the hypotheses and measurement items is positive, then the negative sign of the beta values indicates that the empfeedB ->recomm relationship and the locaNear ->recomm relationship are moving in opposite directions. In other words, the negative sign of the beta indicates that employee feedback and the localNear negatively influence the Recomm. Hence, the rejection of these hypotheses by insignificant p values. For this company, it would be necessary to develop strategies to better satisfy the employees, and to work on the negative points which push them to bring a bad return. Also, the location of the premises (maybe) is an issue. The company should seek to have better geographic coverage.

For hypotheses that have positive beta values and insignificant p values, this indicates that the relationship between these variables and Recomm is positive but not significant. It is up to you to find reasons according to the context of your study to explain why these relationships are not significant. It may be that individuals positively react or accept storArr, CusGree, and Revisit but for some reasons or others, this does not influence them to recommend.

I hope I have given you some answers. Have a nice day!

As Arielle suggested, it would be good if you provide the hypotheses of those two outcomes. Once that is done, people would be able to assist with interpretation of those results. Obviously, the p-values being greater than 0.05 indicates non-significant results. But at this point one is unable to infer anything since we do not know how the hypotheses were formulated. Hope this helps.

The diagram you show is a basic regression equation with 1 dependent variable and 9 independent variables. Since each of these variables is a "partial coefficient," controlling for the other variables in the equation, it is hardly surprising that some of the values would now be non-significant. Also, non-significant effects are not different from zero, so you should not worry about whether they are positive or negative; instead, simply treat them as zero.

The hypotheses I was trying to test is which factors influence customer sastisfaction and how a customer would likely make a recommendation to another collegue or freind to visit the store location

What I think explain the beta negative values is that many customers that answer my survey and I interacted with them in the store did not live any were near the store location they lived farther away or very far while others lived close by So I believe that is why its negative. I believe the employee feedback also came back negative because many customers never really interacted or asked help from the employees within the store because they never felt the need to so they answered that question on the lower 1-7 rating like 4,3,1,5 many customers don't every ask the employees for help and I believe that is due to the way the store is arranged everything is in view and you can always find the item you are looking for

With these things in mind how do I intpret my Beta and p-values now?

Hello Jay

I would like to provide more orientations.

First, about this: "The hypotheses I was trying to test is which factors influence customer satisfaction and how a customer would likely make a recommendation to another colleague or friend to visit the store location". This is not a hypothesis but a research question. According to your research model, the main variable you are studying here is Recommendation. So, why are you talking about customer satisfaction in your research question? You are not studying it. Your research model seems to have this research question: What factors influence a customer to make a recommendation to other colleagues or friends to visit the store location?

Second, about your questionnaire, I am doubtful. You talk about SEM model. With SEM methodology (CB-SEM or PLS-SEM), you have to follow specific steps. The way you structure your questionnaire does not seem to be right. (1) the measurement scale you use for each variable is not the same. If you choose to use a 7-point scale, do so for all variables. At this point, your results for the measurement model criteria (Cronbach, CR, rho) might be above 0.95 or close to 1. This is not good.

Third, how do you theorize your study? the literature review provides you with items that measure well the variable you are studying.

Find a few articles below. Look at how they define the research question. How they formulate the hypothesis. How they discuss the results:

Article In-store consumer behavior: How mobile recommendation agents...

Article Ten Steps in Scale Development and Reporting: A Guide for Researchers

Hello Jay

I'm so disappointed because I don't know why your professor valid your questionnaire. So I can't say more because I don't know your teacher plan for this course.

I just recommend you to look for some article about PLS SEM on Google scholar. Read about the methodology, questionnaire design, and interpretation of the results. Look for some article close to your subject and see how the authors conducted their study. Do it by your own and try to understand.

https://doi.org/10.1108/EBR-11-2018-0203

Please, it is not good to say that you are not smart. Every person is endowed with intelligence. If you have difficulties to understand concepts, it is not because of lack of intelligence but because of lack of method. I don't know you but through your concerns, I see a curious person, very motivated who wants to understand and doesn't hesitate to engage in this study. These are essential characteristics of a good researcher. It is with the mistakes that we learn and that we become better. The questions you ask are not stupid. Just know that information is available to everyone today. Don't just look at what is given in the course, but read the publications yourself. At the beginning it will be a bit difficult if you are not used to it but it will come with perseverance. Don't get stuck because you think you are not smart. This is a very bad thought that is actually very wrong. Don't let your environment and maybe the people you work with make you accept this wrong idea. Keep working and you will get your results.

Concerning your questionnaire.

I understand that your research question is: what are the factors that influence customers to recommend a retailing store to their friends and family? So your study is focused on the concept of "Customer Recommendation" or "Intention to recommend".

1- Go to the literature to identify the research that has been done on this concept. The factors that have been found by the researchers to motivate customer recommendation and the theories used.

2- Let's say hypothetically that your literature review reveals that product quality, product cost or employee empathy are among the determining factors. For each of these variables, you must look for measurement items in the research articles. The measurement items are statements that individuals answer according to a chosen Likert scale. In the image below, here is an example of measurement items (Indicators). Reference: Ukpabi, D.C., Karjaluoto, H., Olaleye, S., Mogaji, E. (2020). Customer Value Framework and Recommendation Intention: The Moderating Role of Customer Characteristics in an Online Travel Community. In: Neidhardt, J., Wörndl, W. (eds) Information and Communication Technologies in Tourism 2020. Springer, Cham. https://doi.org/10.1007/978-3-030-36737-4_4

Design a table with three columns: Variables, measurement items, and the reference of the article where the measurement items were taken.

Then adapt each item to your study context. For example:

Here is an original item: I would recommend this online travel community to friends

Here is the adapted item: I would recommend this retailing store to friends

3- After identifying and adapting all items. Set up your questionnaire in Microsoft Word. First, introduce the study, its objectives and provide assurance of data confidentiality. Second, insert the measurement items, third, apply a 7-point likert scale as a response mode. For example, Strongly Disagree =1, Moderately Disagree =2, Slightly Disagree =3, Neutral =4, Slightly Agree =5, Moderately Agree =6, Strongly Agree =7. Fourth, ask questions about age, gender, frequency of store visits, etc.

4- pretest your questionnaire

5 - Determine the sample size of your study using the Gpower software. Insert the number of independant variables then calculate to have the total sample size. Find orientation in the image i provided

6- Conduct a pilot test with a sample of 50 responses to ensure that the measurement model is correct

7- Recollect your data from the customers of the store, this time trying to involve your co-workers.

8- Read, read and read about PLS SEM methodology!!!

Am I the only one here who knows how to interpret a simple regression? Your effects are not negative, instead they are effectively zero.