Management Research

Research Methods: Discrete Choice Experiment - Number of Selections

Hello everyone,

Today, I respond to another group of comments of my listeners. I try to explain how many decision options you should/could have for each decision within your discrete choice experiment. Later I noticed that I confused the words "decision", "decision option", "selection", or "object". Here, I try to clarify the terminology:

1. Within a DCE you address a certain object (e.g. a car)

2. You ask participants to make a certain number of decisions (e.g. select among two cars). 

3. In one such decision participants are offered several decision options (e.g. that car in red and the car in blue).

4. If the selection is made, the selected option is coded as 1, the neglected option is coded as zero.

5. You can have an option "None of the proposed options". If selected, this option is coded as one and all neglected options are coded as zeros.

6. Both selected and neglected options are included in the analysis as data cases, as a person may indicate a preference towards the selected option or disfavour against the neglected options.

Sorry for the mistakes when talking, now it should be clarified!

 

Best

Eugene (Yevgen)

Research Methods: Discrete Choice Experiment - Research Models with DCE

Hello everyone

 

Today, I respond to some of the comments by my listeners. I have been asked several times about what kind of research models one can approach using a DCE. Moreover, my students ask me about dependent and independent variables. In this episode, I explain what are the dependent and independent variables in a DCE as well as the moderating variables. Please note, that if you want to test interaction(s) (or moderation) with your DCE, your sample size should be larger.

Enjoy, subscribe, and comment!

 

Best,

Eugene (Yevgen)

Research Methods: Discrete Choice Experiment - Attributes and Theory

Good morning!

Today, I respond to some of the requests from my students, colleagues, and podcast listeners. After a series of podcasts and YoutTube videos dedicated to the Discrete Choice Experiment, I decided to add some new episodes addressing these requests one by one. Today, I explain how to derive your attributes from your theory.

Good luck!

Best

Eugene (Yevgen)

Research Methods: Biases in Research - Common Method Bias

Welcome! Today we talk about a very important bias - the common method bias. It can happen that your research method itself, e.g. your questionnaire, creates a skewness in your answers. For instance, if your Likert scales have different font sizes or if the checkboxes on your Likert have different sizes, you may prime your respondents and (un)internationally force them to select a specific answer option. 

In this episode, I explain how the bias appears, how to avoid it, and how to check whether your study contains a common method bias. Enjoy!

Best,

Eugene (Yevgen)

Research Methods: Biases in Research - Response Bias

Welcome! Last time we talked about the non-response bias. Today, we talk about the opposite - the response bias. Unfortunately, some people are motivated to participate in a survey because of, e.g. their anger or love. Angry customers are always ready to share their experience, whereas those who are satisfied are indifferent or postpone your survey. Your family members are always responding to all your surveys (because they love you). Yet, being unconsciously obliged to respond, they are not obliged to tell truth or even invest efforts. Thus, those who respond may skew your results even more than those who ignore your survey because of e.g. bad design. It is interesting, isn't it?

Enjoy the new episode!

Best

Eugene (Yevgen) 

Research Methods: Biases in Research - Non-Response Bias

Morning! Today I talk about one of the most important biases in survey research - the non-response bias. I present you an explanation, some examples, as well as I give you some ideas about how to test for the non-response bias. 

Enjoy!

Best

Eugene (Yevgen)

Research Methods: Biases in Research - Sample Selection Bias

I think we are slowly approximating the end of the topic of biases. No, please do not think, the biases are over - a friend and colleague of mine Professor Ganna Pogrebna wrote a book on biases in research and, believe me, there are many more. So feel free to google this book or write in comments if you haven't managed to find it. In this podcast, I mentioned only the main biases I encountered during my research. 

Today, we talk about sample selection bias. It is dangerous in research but even more dangerous is this bias in politics when election polls are conducted. Listen carefully and try to avoid this bias in your research.

Good luck!

Eugene (Yevgen)

Research Methods: Biases in Research - Omitted Variable Bias

Welcome to my new podcast episode. Today, I talk about the "omitted variable bias". This bias is one of the most impactful for your research and, unfortunately, there is no universal cure apart from reading as much as you can. Of course, some researchers would argue that this is not a bias but a "normal theory development" - we find a relationship that we cannot explain. And this motivates us to look further until we find the "missing variable".  I would agree if we talk about new phenomena. Yet, if a phenomenon has already been investigated and an important variable is not included because of not knowing about this research or with intent, then we talk about the omitted variable bias.

By the way, have you ever thought about it: many populist politicians use the omitted variable bias exactly to draw attention to the wrong relationships and "somehow forget" to report the true reasons for the issue?

Enjoy!

Best,

Eugene (Yevgen)

Research Methods: Biases in Research - Wording Bias

Today, we talk about wording bias. Doesn't matter how you ask your question, you influence the perception of the respondent through your words. "Are you satisfied with our services?" implies satisfaction. "In how much are you satisfied or dissatisfied with our services?" is better, although there might be some priming through the "satisfied" used first. "Isn't it exactly what you wanted to have?" is a highly skewing question since the answer "No" would make the person feel uncomfortable or even evoke the feeling of cognitive dissonance or guilt. 

In this episode, I give some hints on how to avoid or mitigate wording bias. Enjoy, like, and subscribe, if you are passionate about research.

 

Best

Eugene (Yevgen)

Research Methods: Biases in Research - Question Order Bias

Hello in the New Year!

I hope you had a niche Holidays time and now you are ready to work on your research again! Today, I introduce a topic on the question order bias. As you as a researcher are responsible for the question order, this bias is a researcher bias. I explain how and why it appears as well as I give some ideas on how to eliminate (or at least mitigate) this bias. Enjoy and feel free to subscribe and comment!

 

Best

Eugene (Yevgen)