Management Research

This week, we talk about cultural bias. This bias is very manifold, and I talk only about a few aspects of this bias. If you want to know more, you will need to study literature in this regard. I hope that some examples from my research may help you avoid similar mistakes in your future studies.

Good luck and Enjoy!

Best

Eugene (Yevgen)

Today, we start talking about biases on the side of the researcher. The topic of our today's episode is the "confirmation bias". This bias appears if a researcher instead of trying to reject his or her hypothesis tries to support it. Apart from being a wrong scientific approach, a "belief" of a researcher may impact his or her formulation of questions or design of the experiment. For instance, the questions which could reject a hypothesis are forgotten/omitted. Listen more about the bias in the current episode!

Best

Eugene (Yevgen)

Welcome! How did you feel today? How did you feel last month? Asking such questions evoke the recall bias - instead of relying on (real) memory, we "recall", i.e. we construct the memory. Here, our current feelings, intuition, or state may skew our answers. This bias is called "recall bias". More about this bias in the current episode.

Enjoy!

Faithfully yours

Eugene (Yevgen)

Welcome to the next episode in my mini-series on biases in research! Today we talk about two (as to me, related) biases: Acquiescence bias, i.e. a tendency to agree with the interviewer; and the Sponsor bias, i.e. the (unconscious) tendency to give answers, the respondent assume the sponsor of the survey would like to hear.

Enjoy and leave your comments!

Best,

Eugene (Yevgen)

Have you ever noticed when answering a series of questions, all relating to the same topic, that your answers correlate? Actually, they should as usually you are asked about different aspects of the same latent constructs. Yet, sometimes we tend to set the mark without a specific intent - you just mark everything slightly to the right side (or to the left one). We do not go too deep into the aspects and answer with some degree of automaticity. 

Not sure? Just our one reversely coded item and you will notice that even reversely coded it does not show a full negative correlation. At least in my experience, it is the case. More on the issue in this episode. Enjoy!

 

Best

Eugene (Yevgen)

Welcome to my new mini-series on biases in research. Today I talk about one of the "Respondent biases". In particular, I talk about social desirability. First, I explain the issue. Then, I give you some hints on how to test it. Later, I give you advice on how to avoid social desirability. Some examples follow.

Good luck with your research!

 

Best

Eugene (Yevgen)

Morning!

 

And here comes our final episode in a mini-series on research models. Please let me know if you have a specific question which I should add to this mini-series. 

In this episode, I talk about more complex models and an unlimited number of combinations of moderating and meditation effects you can add to your model. Yet, my recommendation is: remain as simple as possible but not simpler.

Good luck!

Faithfully,

Eugene (Yevgen)

Hello everyone

 

After a short break, I continue with the next episode in the mini-series on research models. Today, I talk about the moderated moderation, i.e. model incorporating several conditions shifting your main effect between the independent and dependent variables. 

 

Enjoy, subscribe, and share!

 

Best,

Eugene (Yevgen)

The double mediation model is a more complex version of the simple mediation model. In this model, you ask yourself the question "why?" twice. The double mediation model gives deeper into the nature of the hypothesised relationship. Yet, it requires also a deeper theoretical analysis as well as more complex statistical calculations.

Enjoy this episode with a simple example!

Best,

Eugene (Yevgen)

As a researcher, have you ever asked yourself a question "Under which conditions is the hypothesised relationship weaker/stronger?". If yes, then you have already dealt with moderation. A moderator is a variable that shifts the relationship between the independent and dependent variables. Enjoy my example with ice cream and the level of happiness!

Best,

Eugene (Yevgen)