How do I do a research paper

6 steps to successful research

Does the thought of having to carry out your own research project make you sweat? You can take a deep breath, empiricism is much easier than its reputation suggests. Because: Research is like cooking a four-course menu: It depends on adding the right ingredients at the right time. And just like with food, you can't know how it tasted until you have prepared and eaten it.

Structure is everything: research projects need red threads

A common thread is really important for every research project: All phases of the empirical survey are attached to it one after the other - from the first research idea to the presentation of the results and the summary. This applies to surveys as well as to focus groups, content analyzes, expert interviews or scientific observations. Well, one after the other and every research and statistics mess is a thing of the past.

Step 1: The cognitive interest

You can only research if you know exactly what you want to research. That needs to be clarified. Every research project initially has a general interest in knowledge. Research questions or hypotheses are then derived from this. This is done in business from factual contexts, in science in the course of in-depth literature research. The research questions formulate the exact details that are to be examined.

A little tip: It is better to ask: »What influence do the ingredients of a dish have on the taste?« Instead of: »Do the ingredients have an influence?« - otherwise you can only answer your research question with yes or no.

The difference between research question and hypothesis

Hypotheses, in turn, use findings from the literature, from preliminary studies or parallel surveys and make an assumption based on them. There you then formulate: "The more cream is added, the better it tastes" or: "When cream is added, the taste changes". Only when you have precisely worked out your research questions or hypotheses can you think about how you can most efficiently achieve results.

Note: You answer research questions, hypotheses are tested.

Step 2: the setting

You can either think about how you set up your investigation after you have set up the research questions or hypotheses or in parallel. It is better to be precise here: Determining the right research methodology guides your entire project, including evaluation and presentation of results.

Research qualitatively or quantitatively

You can proceed qualitatively - in search of verbal content - or quantitatively - in search of numbers.

Definition of qualitative research: In qualitative settings, you look for verbal answers to questions such as "What must a dish contain in order for it to taste particularly good?"

Definition of quantitative research: Quantitative empirical research, on the other hand, aims to collect numbers in the form of percentages, mean values ​​or the like. For example, you could ask: "Did you like this dish?" Or "Please give this dish a grade."

Step 3: The population

If you don't think carefully about who or where you want to find out something, it won't work. In interaction with the setting of a survey stands the exact definition of your population: Who - in a survey - or where exactly - in an observation or content analysis - should be examined at all? Unfortunately, many people make the mistake of not making this definition at all or not with sufficient precision. This leads to results that cannot be used or are difficult to use because it cannot be determined for what they apply at all. If you can't get red beets in Munich, you can't say that there aren't any in Germany.

Full census or sample?

Once you have clearly defined the population, you can consider whether a full survey - of the entire population - is possible. If this is not the case, you take a sample and examine only part of the population. You choose this part either randomly, according to certain characteristics or at random - depending on what your setting requires.

Step 4: the survey

Now it’s the actual survey using a questionnaire, guide, coding scheme or protocol sheet. Your questions or survey dimensions must be precisely tailored here. This means that there must be no contradictions or imprecision to your research questions or hypotheses. The measurement level is also decisive for how much you can read out of the data later. For example, if you simply ask: "Did you like this dish?", You can read far less from the results than if you asked "Please give this dish a grade" or "How often a year would you want to eat this dish ? ”You formulate.

Step 5: pretest and evaluation

During the pretest, the survey instrument is checked for its suitability. If it does not work satisfactorily, it has to be revised again. After the survey, it's time to evaluate.

Note: data analyzes are initially carried out technically. Only then do you interpret.

Step 6: Collect, evaluate and interpret data

When collecting the data, you check whether you have been able to collect enough data: Did the full survey really collect data in full or did the sample show a satisfactory response? - Otherwise your results may not be representative. If the data are then available, their technical evaluation follows. Here, it is best to use a statistical evaluation program such as SPSS - especially for quantitative surveys. This, too, is easy to learn and use with instructions for use.

Result part: first evaluate, then discuss

When evaluating, you first describe your results in descriptive statistics, then you conclude your population with random samples - the significance tests. Now you can start interpreting the results, i.e. answering your research questions or testing your hypotheses. To do this, you will carry out in-depth data analyzes if necessary. Finished?

Now it is still a matter of preparing and presenting the results as strikingly as possible so that everyone - including those who hear about your examination for the first time - can quickly find their way around. Not that difficult, is it?