**How to formulate a good research question**

The research question is the most important part of the research. A correct question helps you to conduct your research in an easy and flexible way.

In this paper, I will show you how to formulate a correct research question. In addition, I will show you why certain research questions are not good and how they can be improved.

If you still have to start your research, this is the best investment you can make. I'm sure this will prevent you from feeling stupid, or experiencing feelings of abandonment when your research is a mess.

Get acces to this paper. It's free if you use this **discount code: ****free trial**.

Wouldn't it be nice to display an overview of the total research in a single graph? Well, this is possible with a graph called conceptual model.

Creating a research design or a conceptual model is bound by specific rules. In this paper you will learn these rules.

Once you know how to make such a graph, every study becomes a lot easier to work it out. The design gives guidelines for sampling, collecting the data and analyzing the collected data. It is the methodology in a nutshell to find an answer to the research question.

Any research will go much faster or is much easier to understand with a research design or a conceptual model in your mind.

**Nine ways to draw a sample**

I have read many research reports in which it was completely unclear how the sample was drawn. Most researchers just do something what they think is right. Usually, the reason for this behaviour is missing. Nevertheless, the selection of a good sample is crucial. It determines the quality of your research and with a poor sample, the research loses a lot of quality.

Nine forms of sampling can be distinguished. If you apply the correct method correctly in a given situation, you will get a representative response. Then you will have done well. If you do it wrong, your research can become meaningless.

In this paper I explain which methods are available and how to apply them.

Many researchers claim to have a representative response when their sample is large enough. Two mistakes can already be found in this vision. First, the sample size is not related to representativeness. And second, the size of the sample is not the problem. The problem that counts is the size of the response.

Now, you might be worried. Maybe you should be.

Another often used way to calculate the sample size is to calculate an estimated proportion with a confidence interval of 95%. Based on the standard calculations, the result is always a sample size of 384.1 or 385. This view also does not stand in a critical discussion. In many studies, estimating a proportion is not important. It is therefore not a reason to calculate the sample size based on the estimation of a proportion.

The correct way to calculate the sample size is much more complex. And by the way, the right way does not exist.

This paper explains how to deal with the problem of sample size. Read it and discover how you can apply the solution to your research.

**Representativity**

An important topic in research is representativity. Perhaps it is even the most important topic. In the case of a non-representative response, none of the conclusions that can be drawn may be justified.

Different views on representativity can be distinguished. One of them is taking a random sample. It can be demonstrated that with random sampling at least five percent of the samples will be non-representative.

Another view that claims to have a representative response is that the sample is large. However, a representative sample is not the problem. A representative response is required. A non-representative sample can even be used as a strategy to get a representative response.

These and some other views are explained and discussed in this paper.

If you want to perform high quality research, make sure you have a representative response. You can learn how to achieve this in this paper.

**Factor analysis and Cronbach's alpha**

Measuring in science means providing valid and reliable data.

One way to get valid and reliable data is to create a questionnaire with a list of statements about a topic. In questionnaires, respondents are asked to give their answer on a five-point Likert scale. The list of statements is supposed to measure a construct. To statistically test whether the construct has been measured correctly, a form of factor analysis is performed. And to test whether the construct has been reliably measured, Cronbach's alpha must be calculated.

This, in a nutshell, is the essence of why factor analysis should be applied and Cronbach's alpha should be calculated. Much more details can be found in our paper.

**The statistical test procedure**

Statistics are essentially very simple. It is about calculating a value from the data and comparing this value with a value in a table. Is that all? Yes that's all.

However, you must know how to calculate the value from your data and you must know with which value it has to be compared. For this, a procedure must be followed.

This procedure is described in detail in this paper. Some steps in the procedure are simple, others are more difficult and need more thinking time. In the beginning, it might look impossible to learn, however, when you become more experienced, you will be able to complete the whole procedure in a split second.

**How to choose the correct statistical test**

One of the most difficult things in statistics is to figure out which statistical test has to be applied. Before you make a choice, you need to know what variables are, how they are measured, what dependent and independent variables are and whether it is about comparing features or comparing groups.

This papers explains what you need to know about variables and then presents tables to determine which statistical test should be used.

**How to present statistical results**

When you have analyzed the data, a new problem arises: How to present the results in a report?

It is not permitted to copy the output of statistical software programs and paste it one on one in a paper. In fact, an estimated ninety-nine percent of the numbers in the output can be deleted. Most of the time, only a few digits are needed.

A distinction must be made between presenting the results in text and presenting the data in tables. Slightly fewer numbers are presented in tables.

This paper provides a guideline for presenting statistical results. When the rules in this paper are followed, fine-looking tables are produced that are consistent with the rules of many scientific journals.

**Analyzing data from interviews**

Do you get stuck in your research because you don't know what to do with the data from the interviewees? If so, you are not the only one.

Many students think that analyzing interviews is easy. Often they are disappointed. Usually, it takes a lot of work to organize the data in such a way before they can begin the real analysis and write a report on the results of the interviews.

This paper contains a step-by-step plan that explains how you can professionally analyse the data from interviews. Processing the data this way, will prevent bias and you will be able to give a good answer to your research question. It will still be a lot of work, but you will get much better results, so that your research becomes credible.

Because:

Good research provides you with better information.

With better information, you can make better decisions.

With better decisions, you can create a healthier, wealthier and freer world,

for people, fauna and flora, for current and future generations.

That is why I think it is necessary for you to know how to perform research well.