A complete guide to dissertation primary research

For many students, including those who are at postgraduate level or well-versed in the dissertation process, the prospect of primary research can be somewhat daunting.

But we are here to help quell those pesky nerves you may be feeling if you are faced with doing primary research for your dissertation.

Now, this is an ultimate guide and therefore contains plenty of information. To make it easier to digest and jump between sections, here's a summary of its contents:

(Note: Steps 1 and 2, and 9 and 10, apply no matter which method of data collection you choose.)

The reasons that students can feel so wary of primary research can be many-fold. From a lack of knowledge of primary research methods, to a loathing for statistics, or an absence of the sufficient skills required… The apprehension that students can feel towards primary research for their dissertation is often comparable to the almost insurmountable levels of stress before exams.

And yet, there’s a significant difference between doing primary research and sitting exams. The former is far more engaging, rewarding, varied, and dare we say it, even fun. You’re in the driving seat and you get to ask the questions. What’s more, students carrying out primary research have an opportunity to make small contributions to their field, which can feel really satisfying – for many, it’s their first taste of being a researcher, rather than just a learner.

Now, if you’re reading this and scoffing at our steadfast enthusiasm for primary research, we’ll let you in on a little secret – doing research actually isn’t that difficult. It’s a case of learning to follow specific procedures and knowing when to make particular decisions.

Which is where this guide comes in; it offers step-by-step advice on these procedures and decisions, so you can use it to support you both before and during your dissertation research process.

As set out below, there are different primary research methodologies that you can choose from. The first two steps are the same for whatever method you choose; after that, the steps you take depend on the methodology you have chosen.

The first step of your primary research is to decide on the type of data that your study will use. In general, you can select from primary data, secondary data, and big data. Note that this guide focuses on primary research – that is, on research that uses primary data. Nonetheless, it is useful to know a little about each of these data approaches.

Primary data has been collected by the researcher himself or herself. When doing primary research for your undergraduate or graduate degrees, you will most commonly rely on this type of data. It is often said that primary data is “real time” data, meaning that it has been collected at the time of the research project. Here, the data collection is under the direct control of the researcher.

Secondary data has been collected by somebody else in the past and is usually accessible via past researchers, government sources, and various online and offline records. This type of data is referred to as “past data”, because it has been collected in the past. Using secondary data is relatively easy since you do not have to collect any data yourself. However, it is not always certain that secondary data would be 100% relevant for your project, since it was collected with a different research question in mind. Furthermore, there may be questions over the accuracy of secondary data.

Big data is the most complex type of data, which is why it is almost never used during undergraduate or graduate studies. Big data is characterised by three Vs: high volume of data, wide variety of the type of data, and high velocity with which the data is processed. Due to the complexity of big data, standard data processing procedures do not apply and you would need intense training to learn how to process it.

When doing primary research, you have a choice of relying on qualitative, quantitative, or mixed methodologies. Each of these will be described separately, and then we will look at specific steps that you need to complete depending on your chosen methodology.

Qualitative research is exploratory in nature. This means that qualitative research is often conducted when there are no quantitative investigations on the topic, and you are seeking to explore the topic for the first time. This exploration is achieved by considering the perspectives of specific individuals. You are concerned with particular meanings that reflect a dynamic (rather than fixed) reality. By observing or interviewing people, you can come to an understanding of their own perception of reality.

Quantitative research is confirmatory in nature. Thus, the main goal is to confirm or disconfirm hypotheses by relying on statistical analyses. In quantitative research, you will be concerned with numerical data that reflects a fixed and measurable (rather than dynamic) reality. Using large samples and testing your participants through reliable tools, you are seeking to generalise your findings to the broader population.

Mixed research combines qualitative and quantitative methodologies. The goal is to gain a more thorough understanding of a topic than would be possible by relying on a single methodological approach. Usually, a mixed method involves doing qualitative research first, which is then supplemented by quantitative research.

Thus, first you explore a phenomenon through a low-scale study that focuses on the meanings of particular individuals, and then you seek to form a hypothesis and test it with a larger sample. You can rely on other mixed methodologies as well, which will be described later.

One of the main strengths of qualitative research is that you can explore a topic in detail and in depth. This is because you train your magnifying glass on particular people or events. By focusing on this in-depth view, you can understand specific meanings, occurrences, behaviours, thoughts, and circumstances associated with your research topic. The information obtained in qualitative research is far more detailed than in quantitative research.

Still, there are many things that you cannot achieve with qualitative research. Since you are investigating a select group of people, you cannot generalise your findings to the broader population. With qualitative methodology, it is also harder to establish the quality of research. Here, the quality depends on your own skills and your ability to avoid bias while interpreting findings. Finally, qualitative research involves interpreting multiple perspectives, which makes it harder to reach a consensus and establish a “bottom line” of your results and conclusions.

There are several types of qualitative methods and you need to decide which one is most relevant to your research question. The most common qualitative methods are observations, interviews, focus groups, and case studies. Let’s address each of these separately.

An observation, as its name implies, involves observing a group of individuals within their own setting. As a researcher, your role is to immerse yourself in this setting and observe the behaviour of interest. You may either become involved with your participants, therefore taking a participating role, or you can act as a bystander and observer. You will need to complete an observation checklist, which you will have made in advance, on which you will note how your participants behaved.

Interviews are the most common qualitative method. In interviews, your role is to rely on predetermined questions that explore participants’ understanding, opinions, attitudes, feelings, and the like. Depending on your research topic, you may also deviate from your question structure to explore whatever appears relevant in the present moment. The main aim is always to understand participants’ own subjective perspectives.

Focus groups are like interviews, except that they are conducted with more than one person. Participants in a focus group are several individuals usually from different demographic backgrounds. Your role is to engage them in an open discussion and to explore their understanding of a topic.

In both interviews and focus groups, you should record your sessions and transcribe them afterwards.

Case studies focus on a particular person or organisation relevant to your research topic. The reason why such individuals and organisations are chosen for the case study is that they offer an interesting, unusual, or particularly revealing set of circumstances. Your role is to explore how the case study informs your investigation.

When selecting participants, first you need to ask yourself who your participants are. As noted above, your participants will be a group of people when conducting observations, interviews, and focus groups, and a single person or organisation when undertaking a case study.

To establish who these individuals (or organisation) will be, you must consult your research question. You also need to decide on the number of participants, which is generally low in qualitative research, and to decide whether participants should be from similar or different backgrounds. Ask yourself: What needs to connect them and what needs to differentiate them?

Selecting your participants involves deciding how you will recruit them. An observation study, usually has a predetermined context in which your participants operate, and this makes recruitment straightforward since you know where your participants are. A case study does not necessitate recruitment, since your subject of investigation is a predefined person or organisation.

However, interviews and focus groups need a more elaborate approach to recruitment as your participants need to represent a certain target group. A common recruitment method here is the snowball technique, where your existing participants themselves recruit more participants. Otherwise, you need to identify what connects your participants and try to recruit them at a location where they all operate.

Also remember that your participants always need to sign an informed consent, therefore agreeing to take part in the research (see Step 9 below for more information about informed consent and research ethics).

Qualitative research involves the use of measures, although these measures are much simpler than those used in quantitative research.

Observation studies usually involve a checklist on which you note your observations. Focus groups, interviews, and most case studies use structured or semi-structured interviews. Structured interviews rely on a predetermined set of questions, whereas semi-structured interviews combine predetermined questions with an opportunity to explore the responses further.

In most circumstances, you will want to use semi-structured interviews because they enable you to elicit more detailed responses.

When using observations, you will need to craft your observation checklist; when using interviews, you will need to craft your interview questions. Observation checklists are easy to create, since they involve your predetermined focuses of observation. These may include participants’ names (or numbers if they are to remain anonymous), their demographic characteristics, and whatever it is that you are observing.

When crafting interview questions, you will need to consult your research question, and sometimes also the relevant literature. For instance, if your interviews focus on the motivations for playing computer games, you can craft general questions that relate to your research question – such as, “what motivates you to play computer games the most?”.

However, by searching the relevant literature, you may discover that some people play games because they feel competent, and then decide to ask your subjects, “does playing games satisfy your need for competence?”

No matter what informs your creation of an observation checklist or interview questions, you should always consult your supervisor for advice.

When analysing the results of qualitative research, you don’t need to rely on complex statistics. This is because you will not have used any numerical data. (Note, however, that you do need to calculate the mean age of participants, and the percentage distribution of participants according to gender, educational background, ethnicity, and the like).

With qualitative research, your data analysis relies on coding and finding themes in your data.

The process of coding should be straightforward. You simply need to read through your observation checklist or interview transcripts and underline each interesting observation or answer.

Finding themes in your data involves grouping the coded observations/answers into patterns. This is a simple process of grouping codes, which may require some time, but is usually interesting work. There are different procedures to choose from while coding and finding themes in your data. These include thematic analysis, interpretative phenomenological analysis, constant comparative analysis, narrative analysis, and discourse analysis. Let’s define each of these.

Thematic analysis is the most commonly used procedure. The goal is to code data, search for themes among the codes, review themes, and name themes. Interpretative phenomenological analysis is done in the same way as a thematic analysis, but with a special focus on understanding how a given person, in a given context, makes sense of a phenomenon.

In constant comparative analysis, one piece of data (one theme, one statement) is compared to all others to detect similarities or differences; this comparison aims at understanding the relations between the data. Narrative analysis also extracts themes, but it focuses on the way that participants represent their experiences linguistically.

Finally, discourse analysis also focuses on language, but the goal is to understand how language relates to the influences that shape people’s thoughts and behaviours.

The procedure of conducting qualitative research is quite straightforward. Once you have recruited your participants, you need to schedule the time when you will observe them (in the case of observations) or interview them (in the case of interviews, focus groups, and case studies).

During observations and interviews, you are collecting your data. As noted previously, you always need to record your interviews and transcribe them afterwards. Observation checklists merely need to be completed during observations.

Once your data has been collected, you can then analyse your results, after which you should be ready to write your final report.

Quantitative research has many advantages. By relying on well-developed participant selection procedures, you can select a sample that is representative of the population under investigation. This, together with the use of statistical analysis, helps you to generalise your results to the given population.

Quantitative research also enables you to test hypotheses and to determine causality. Determining causality is possible because a quantitative method allows you to control for extraneous variables (confounders) that may affect the relationship between certain variables. Finally, by using standardised procedures, your quantitative study can be replicated in the future, either by you or by other researchers.

At the same time, however, quantitative research also has certain limitations. For instance, this type of research is not as effective at understanding in-depth perceptions of people, simply because it seeks to average their responses and get a “bottom line” of their answers.

In addition, quantitative research often uses self-report measures and there can never be certainty that participants were honest. Sometimes, quantitative research does not include enough contextual information for the interpretation of results. Finally, a failure to correctly select your participants, measures and analyses might lower the generalisability and accuracy of your findings.

There are different quantitative methods to choose from, and your choice will depend on your research question. In general, you can select from doing descriptive, correlational, experimental, or quasi-experimental research.

Let’s look at each of these separately.

Descriptive research is used when you want to describe characteristics of a population or a phenomenon. For instance, if you want to describe how many college students use drugs and which drugs are most commonly used, then you can use descriptive research.

You are not seeking to establish the relationship between different variables, but merely to describe the phenomenon in question. Thus, descriptive research is never used to establish causation.

Correlational research is used when you are investigating a relationship between two or more variables. The concept of independent and dependent variables is important for correlational research.

An independent variable is one that you control to test the effects on the dependent variable. For example, if you want to see how intelligence affects people’s critical thinking, intelligence is your independent variable and critical thinking is your dependent variable.

In scientific jargon, correlation tests whether an independent variable relates to the levels of the dependent variable(s). Note that correlation never establishes causation – it merely tests a relationship between variables.

You can also control for the effect of a third variable, which is called a covariate or a confounder. For instance, you may want to see if intelligence relates to critical thinking after controlling for people’s abstract reasoning.

The reason why you might want to control for this variable is that abstract reasoning is related to both intelligence and critical thinking. Thus, you are trying to specify a more direct relationship between intelligence and critical thinking by removing the variable that “meddles in between”.

Experiments aim to establish causation. This is what differentiates them from descriptive and correlational research. To establish causation, experiments manipulate the independent variable.

Or, to put this another way, experiments have two or more conditions of the independent variable and they test their effect on the dependent variable. Here’s an example: You want to test if a new supplement (independent variable) increases people’s concentration (dependent variable). You need a reference point – something to compare this effect to. Thus, you compare the effect of a supplement to a placebo, by giving some of your participants a sugary pill.

Now your independent variable is the type of treatment, with two conditions – supplement and placebo. By comparing the concentration levels (dependent variable) between participants who received the supplement versus placebo (independent variable), you can determine if the supplement caused increased concentration.

Experiments can have two types of designs: between-subjects and within-subjects. The above example illustrated a between-subjects design because concentration levels were compared between participants who got a supplement and those who got a placebo.

But you can also do a within-subjects comparison. For example, you may want to see if taking a supplement before or after a meal impacts concentration levels differently. Here, your independent variable is the time the supplement is taken (with two conditions: before and after the meal). You then ask the same group of participants to take on Day 1 the supplement before the meal and on Day 2 after the meal.

Since both conditions apply to all your participants, you are making a within-subjects comparison. Regardless of the type of design you are using, when assigning participants to a condition, you need to ensure that you do so randomly.

A quasi-experiment is not a true experiment. It differs from a true experiment because it lacks random assignment to different conditions. You would use a quasi-experiment when your participants are grouped into different conditions according to a predetermined characteristic.

For instance, you may want to see if children are less likely than adolescents to cheat on a test. Here, you are categorising your participants according to their age and thus you cannot use random assignment. Because of this, it is often said that quasi-experiments cannot properly establish causation.

Nonetheless, they are a useful tool for looking at differences between predetermined groups of participants.

Your sample in quantitative research usually consists of a much higher number of participants than in qualitative research. As mentioned previously, having large sample sizes in quantitative research helps you to generalise your findings to the broader population.

A very good practice is to rely on a G-Power analysis to calculate how large your sample size should be in order to increase the accuracy of your findings. G-Power analysis, for which you can download a program online, is based on a consideration of previous studies’ effect sizes, significance levels, and power.

Thus, you will need to find a study that investigated a similar effect, dig up its reported effect size, significance level and power, and enter these parameters in a G-Power analysis. There are many guides online on how to do this.

When selecting participants for your quantitative research, you also need to ensure that they are representative of the target population. You can do this by specifying your inclusion and exclusion criteria.

For instance, if your target population consists of young women who have given birth and who have depression, then you will include only women who have given birth, who are younger than 35, and who have depression. Consequently, you will exclude women who do not fulfil these criteria.

Remember that here, just as in qualitative research, you need informed consent from your participants, therefore ensuring that they have agreed to take part in the research.

Quantitative research is all about using measurement tools. Thus, all variables that you use in your research will be “operationalised” by using specific measures. These measures consist of reliable and valid questionnaires which have been used in past research.

A questionnaire is reliable when it has led to consistent results across studies, and it is valid when it measures what it is supposed to measure. You can claim that a questionnaire is valid and reliable when previous studies have established its validity and reliability (you must cite these studies, of course).

Moreover, you can test the reliability of a questionnaire yourself, by calculating its Cronbach’s Alpha value in a statistics programme, such as SPSS. Values higher than 0.7 indicate acceptable reliability, values higher than 0.8 indicate good reliability, and values higher than 0.9 indicate excellent reliability. Anything below 0.7 indicates unreliability.

You can always consult your supervisor about which questionnaires to use in your study. Alternatively, you can search for questionnaires yourself by looking at previous studies and the kind of measures they employed.

Each questionnaire that you decide to use will require you to calculate final scores. You can obtain the guidelines for calculating final scores in previous studies that used a given questionnaire. You will complete this calculation by relying on a statistical program.

Very often, this calculation will involve reverse-scoring some items. For instance, a questionnaire may ask “are you feeling good today?”, where a response number of 5 means “completely agree”. Then you can have another question that asks “are you feeling bad today?”, where a response number of 5 again means “completely agree”. If your questionnaire measures whether a person feels good, then you will have to reverse-score the second of these questions so that higher responses indicate feeling more (rather than less) good. This can also be done using a statistics program.

Statistical analyses, which are a necessary step of quantitative research, often give students a headache.

However, there is no reason why they should, since the whole procedure of doing statistical analyses is not that difficult – you just need to know which analysis to use for which purpose and to read guidelines on how to do particular analyses (online and in books). Let’s provide specific examples.

If you are doing descriptive research, your analyses will rely on descriptive and/or frequencies statistics.

Descriptive statistics include calculating means and standard deviations for continuous variables, and frequencies statistics include calculating the number and percentage of the frequencies of answers on categorical variables.

Continuous variables are those where final scores have a wide range. For instance, participants’ age is a continuous variable, because the final scores can range from 1 year to 100 years. Here, you calculate a mean and say that your participants were, on average, 37.7 years old (for example).

Another example of a continuous variable are responses from a questionnaire where you need to calculate a final score. For example, if your questionnaire assessed the degree of satisfaction with medical services, on a scale ranging from 1 (not at all) to 5 (completely), and there are ten questions on the questionnaire, you will have a final score for each participant that ranges from 10 to 50. This is a continuous variable and you can calculate the final mean score (and standard deviation) for your whole sample.

Categorical variables are those that do not result in final scores, but result in categorising participants in specific categories. An example of a categorical variable is gender, because your participants are categorised as either male or female. Here, your final report will say something like “50 (50%) participants were male and 50 (50%) were female”.

Please note that you will have to do descriptive and frequencies statistics in all types of quantitative research, even if your research is not descriptive research per se. They are needed when you describe the demographic characteristics of your sample (participants’ age, gender, education level, and the like).

When doing correlational research, you will perform a correlation or a regression analysis. Correlation analysis is done when you want to see if levels of an independent variable relate to the levels of a dependent variable (for example, “is intelligence related to critical thinking?”).

You will need to check if your data is normally distributed – that is, if the histogram that summarises the data has a bell-shaped curve. This can be done by creating a histogram in a statistics program, the guidelines for which you can find online. If you conclude that your data is normally distributed, you will rely on a Pearson correlation analysis; if your data is not normally distributed, you will use a Spearman correlation analysis. You can also include a covariate (such as people’s abstract reasoning) and see if a correlation exists between two variables after controlling for a covariate.

Regression analysis is done when you want to see if levels of an independent variable(s) predict levels of a dependent variable (for example, “does intelligence predict critical thinking?”). Regression is useful because it allows you to control for various confounders simultaneously. Thus, you can investigate if intelligence predicts critical thinking after controlling for participants’ abstract reasoning, age, gender, educational level, and the like. You can find online resources on how to interpret a regression analysis.

When you are conducting experiments and quasi-experiments, you are using t-tests, ANOVA (analysis of variance), or MANCOVA (multivariate analysis of variance).

Independent samples t-tests are used when you have one independent variable with two conditions (such as giving participants a supplement versus a placebo) and one dependent variable (such as concentration levels). This test is called “independent samples” because you have different participants in your two conditions.

As noted above, this is a between-subjects design. Thus, with an independent samples t-test you are seeking to establish if participants who were given a supplement, versus those who were given a placebo, show different concentration levels. If you have a within-subjects design, you will use a paired samples t-test. This test is called “paired” because you compare the same group of participants on two paired conditions (such as taking a supplement before versus after a meal).

Thus, with a paired samples t-test, you are establishing whether concentration levels (dependent variable) at Time 1 (taking a supplement before the meal) are different than at Time 2 (taking a supplement after the meal).

There are two main types of ANOVA analysis. One-way ANOVA is used when you have more than two conditions of an independent variable.

For instance, you would use a one-way ANOVA in a between-subjects design, where you are testing the effects of the type of treatment (independent variable) on concentration levels (dependent variable), while having three conditions of the independent variable, such as supplement (condition 1), placebo (condition 2), and concentration training (condition 3).

Two-way ANOVA, on the other hand, is used when you have more than one independent variable.

For instance, you may want to see if there is an interaction between the type of treatment (independent variable with three conditions: supplement, placebo, and concentration training) and gender (independent variable with two conditions: male and female) on participants’ concentration (dependent variable).

Finally, MANCOVA is used when you have one or more independent variables, but you also have more than one dependent variable.

For example, you would use MANCOVA if you are testing the effect of the type of treatment (independent variable with three conditions: supplement, placebo, and concentration training) on two dependent variables (such as concentration and an ability to remember information correctly).

The procedure of conducting quantitative research is easy.

If you are doing an experiment, once you have recruited your participants, you need to randomly assign them to conditions. If you are doing quasi-experimental research, you will have a specific procedure for predetermining which participant goes to which condition. For instance, if you are comparing children versus adolescents, you will categorise them according to their age. In the case of descriptive and correlational research, you don’t need to categorise your participants.

Furthermore, with all procedures you need to introduce your participants to the research and give them an informed consent. Then you will provide them with the specific measures you are using.

Sometimes, it is good practice to counterbalance the order of questionnaires. This means that some participants will get Questionnaire 1 first, and others will be given Questionnaire 2 first.

Counterbalancing is important to remove the possibility of the “order effects”, whereby the order of the presentation of questionnaires influences results.

At the end of your study, you will “debrief” your participants, meaning that you will explain to them the actual purpose of the research. After doing statistical analysis, you will need to write a final report.

The main strength of mixed research is that it overcomes the specific limitations of qualitative and quantitative research. For instance, qualitative research is limited because it is prone to biases when interpreting results and because results cannot be generalised to the broader population – which is something that quantitative research makes up for.

Quantitative research, on the other hand, is limited because it does not lead to an in-depth understanding of particular meanings and contexts – something that qualitative research makes up for. Thus, by using the mixed method, the strengths of each approach are making up for their respective weaknesses. You can, therefore, obtain more information about your research question than if you relied on a single methodology.

Mixed research has, however, some limitations. One of its main limitations is that research design can be quite complex. It can also take much more time to plan mixed research than to plan qualitative or quantitative research.

Sometimes, you may experience difficulties in bridging your results because you need to combine the results of qualitative and quantitative research. Finally, you may find it difficult to resolve discrepancies that occur when you interpret your results.

For these reasons, mixed research needs to be done and interpreted with care.

There are several types of mixed methods. They are differentiated based on how and when the qualitative and quantitative research elements are done. These methods are called sequential exploratory design, sequential explanatory design, concurrent triangulation design, and concurrent nested design.

Let’s address each of these separately.

Sequential exploratory design is a method whereby qualitative research is done first and quantitative research is done second. By following this order, you can investigate a topic in-depth first, and then supplement it with numerical data. This method is useful if you want to test the elements of a theory that stems from qualitative research and if you want to generalise qualitative findings to different population samples.

Sequential explanatory design is when quantitative research is done first and qualitative research is done second. Here, priority is given to quantitative data. The goal of subsequent collection of qualitative data is to help you interpret the quantitative data. This design is used when you want to engage in an in-depth explanation, interpretation, and contextualisation of quantitative findings. Alternatively, you can use it when you obtain unexpected results from quantitative research, which you then want to clarify through qualitative data.

Concurrent triangulation design involves the simultaneous use of qualitative and quantitative data collection. Here, equal weighting is given to both methods and the analysis of both types of data is done both separately and simultaneously.

This design is used when you want to obtain detailed information about a topic and when you want to cross-validate your findings. Cross-validation is a statistical procedure for estimating the performance of a theoretical model that predicts something. Although you may decide to use concurrent triangulation during your research, you will probably not be asked to cross-validate the findings, since this is a complex procedure.

Concurrent nested design is when you collect qualitative and quantitative data at the same time, but you employ a dominant method (qualitative or quantitative) that nests or embeds the less dominant method (for example, if the dominant method is quantitative, the less dominant would be qualitative).

What this nesting means is that your less dominant method addresses a different research question than that addressed by your dominant method. The results of the two types of methods are then combined in the final work output. Concurrent nested design is the most complex form of mixed designs, which is why you are not expected to use it during your undergraduate or graduate studies, unless you have been specifically asked.

When doing mixed research, you will often have two groups of participants – one that takes part in qualitative research and the other that takes part in quantitative research. Please read the above sections on how to select participants depending on whether the research is qualitative or quantitative.

In summary, participants in the qualitative part of the investigation will be several individuals who are relevant for your research project. Conversely, your sample size for the quantitative part of the investigation will be higher, including many participants chosen as representative of your target population.

You will also need to rely on different recruitment strategies when selecting participants for qualitative versus quantitative research.

Because mixed research combines qualitative and quantitative methodologies, you will need to use two types of measures – those appropriate to qualitative research and those appropriate to quantitative research.

An in-depth explanation of these measures has been provided in the sections above dealing with qualitative and quantitative research respectively.

In summary, qualitative research relies on the use of observations or interviews, which you usually need to craft yourself. Quantitative research relies on the use of reliable and valid questionnaires, which you can take from past research.

Sometimes, within the mixed method, you will be required to craft a questionnaire on the basis of the results of your qualitative research. This is especially likely if you are using the sequential exploratory design, where you seek to validate the results of your qualitative research through subsequent quantitative data.

In any case, a mixed method requires a special emphasis on aligning your qualitative and quantitative measures, so that they address the same topic. You can always consult your supervisor on how to do this.

Because qualitative and quantitative research require different data analysis procedures, a mixed method will require you to do both. You can look above for more information on specific analyses used in these methods.

In general, qualitative investigations require you to thematically analyse your data, which is done through coding participants’ answers or your observations and finding themes among the codes. You can rely on a thematic analysis, interpretative phenomenological analysis, constant comparative analysis, narrative analysis, and discourse analysis.

Quantitative investigations require you to do statistical analyses, the choice of which depends on the type of quantitative design you are using. Thus, you will use descriptive statistics if you do descriptive research, correlation or regression if you are doing correlational research, and a t-test, ANOVA, or MANCOVA if you are doing an experiment or a quasi-experiment.

Your procedure for doing mixed research will largely depend on the type of mixed design that you are using. Thus, if you are using the sequential exploratory design, you will first undertake the qualitative part of the investigation and then the quantitative part of your research.

The reverse is true if you are using the sequential explanatory design. Concurrent triangulation and concurrent nested designs require you to perform the qualitative and quantitative parts of the investigation simultaneously; what differentiates them is whether you are (or are not) prioritising one (either qualitative or quantitative) as a predominant method.

Regardless of your mixed design, you will need to follow specific procedures for qualitative and quantitative research, all of which have been described above.

Whether you are doing a qualitative, quantitative, or mixed study, you need to ensure that your primary research is ethical.

Some studies deal with especially vulnerable groups of people or with sensitive topics. It is extremely important, therefore, that no harm is done to your participants.

Before beginning your primary research, you will submit your research proposal to an ethics committee. Here, you will specify how you will deal with all possible ethical issues that may arise during your study. Even if your primary research is deemed as ethical, you will need to comply with certain rules and conduct to satisfy the ethical requirements. These consist of providing informed consent, ensuring confidentiality and the protection of participants, allowing a possibility to withdraw from the study, and providing debriefing.

Informed consent ensures that participants have understood the research and have agreed to participate. Regardless of the type of research you are doing, you will solicit informed consents. Participants can provide consent by signing a printed or an online consent form. If your participants cannot provide an online signature, you can simply tell them that, by proceeding to the next page of the questionnaire, they are agreeing to take part in the research.

Informed consent needs to be signed by individuals who are older than 18. If your participants are younger, you will need to obtain consent from parents or legal guardians. Finally, for a variety of reasons it may sometimes be impossible to obtain an informed consent from your participants. If so, you can ask a similar group of people how they would feel to take part in your study; if they agree to participate, you have obtained a “presumptive consent”.

Confidentiality of your participants is ensured through keeping all data anonymous. Thus, you will never ask for your participants’ names – instead, you will provide each with a participant number. Even if you are reporting interviews with several participants, you shouldn’t refer to them by names, but by their initials.

It is also important to keep your data safe, so that it cannot be accessed by third parties. Moreover, the importance of protecting your participants means you must ensure that they will not suffer any physical and mental consequences during or after their participation. You shouldn’t embarrass them, frighten them, or offend them. If your participants represent vulnerable groups (children, the elderly, disabled, etc.), you need to provide them with special care during research.

Importantly, your participants should always be informed that they can withdraw from the research at any point. This can be done either during their participation (such as during an interview or a questionnaire) or after their participation. In the latter case, participants should be able to contact you and ask for their data to be destroyed.

Finally, you always need to debrief your participants at the end of the study. This is especially important when you have “deceived” your participants by not telling them the true purpose of your research in advance (so as to remove bias). You can debrief your participants either face-to-face or through providing a typed debriefing form.

Undergraduate studies involve less complex research than graduate studies.

If you’re an undergraduate student, your primary research will usually rely on qualitative or quantitative methodology, rather than on mixed methodology. When doing qualitative research, you will use interviews more often than observations, focus groups, or case studies. You may even get help from your supervisor to craft your interview questions.

When doing quantitative research, you may rely on the full range of methodologies (descriptive, correlational, experimental, and quasi-experimental), but your study will be simple and straightforward. Your sample size won’t be too high, you won’t need to calculate your sample size in advance, you will use only a few measures, and you will use simpler statistical analyses. The main goal of your undergraduate research project is to help you learn the basics of research.

Graduate studies demand more involvement with a research project. If you’re a graduate student, you need to think critically and decide on the best way to answer your research question. Thus, you should consider all the primary research procedures that were mentioned in this guide. You can use a qualitative, quantitative, or mixed methodology.

If doing qualitative research, you need to select the most suitable method, craft your measures, and show an in-depth understanding of thematic analysis (or other similar methods). If doing quantitative research, you need to design your study well, calculate your sample size in advance, recruit a high number of participants, test multiple hypotheses, and rely on more complex statistical analyses. The main goal of your graduate research project is to help you learn more advanced methods of research.

The final thing that you need to know about doing primary research is that certain research methods are more commonly used in some disciplines than in others.

Qualitative research is mostly used in the social sciences, and slightly less in natural and formal (mathematical) sciences. Because of their frequent reliance on qualitative research, social sciences are often called “soft sciences” (which doesn’t make them easier disciplines!)

Observations are used in ethnology and cultural anthropology, but also in sociology, psychology, education, human geography, and communication studies; interviews are used in all disciplines that favour qualitative research; focus groups are used in library and information sciences, social sciences, business studies, and usability engineering; and case studies are used in administrative science, social work, clinical science, education, anthropology, sociology, psychology, and political science.

Quantitative research is favoured in all social sciences (although a little less in geography and anthropology) and it forms the basis of all natural and formal (mathematical) sciences (which is why they are called “hard sciences” – which doesn’t mean they are harder disciplines!). Regardless of the discipline, you need to be familiar with all basic quantitative methods (descriptive, correlational, experimental, and quasi-experimental). These are regarded as cornerstones of science, especially when such methods seek to establish causality.

Finally, mixed research is most favoured within the social sciences, although natural and formal sciences also benefit from it.

This guide has summarised the main steps of conducting primary research for your undergraduate or graduate project. These steps are as follows: