How Do You Know Whether You Have the Skills Needed to Conduct the Study?

Your Study Looks Feasible, but Can You Actually Do What It Requires?

You have a research question, access to participants or data, enough time, and perhaps even the necessary funding. Then you examine the methodology more closely.

The study requires multilevel modeling, but you have never used it. You plan qualitative interviews but have never conducted or analyzed one. The dataset contains millions of records and requires programming rather than a spreadsheet. Your experiment depends on laboratory procedures you have only observed. Or the project requires research software you have never opened.

Does that mean you should choose another study?

Not necessarily. Research is partly a process of developing new capabilities. The more useful question is whether you, your supervisors, collaborators, or specialists available to the project can perform every critical task competently by the time it needs to be done. A skill you do not possess today is a manageable gap if there is a credible route to acquiring or accessing it. A critical skill with no such route is a feasibility problem.

You Need Competence Across the Study, Not Personal Mastery of Everything

How to Determine Whether Your Research Capabilities Match the Study

Start with the tasks the study requires

Asking whether you are "good at research" is too vague to be useful. Research competence is not one skill.

A study consists of specific tasks. Depending on the design, these might include developing an instrument, conducting interviews, recruiting participants, obtaining informed consent, collecting biological specimens, managing sensitive records, programming an experiment, cleaning a complex dataset, coding qualitative material, conducting statistical analysis, interpreting results, or maintaining reproducible analytical files.

List those tasks before assessing your competence. The relevant question is not whether you consider yourself a competent researcher in general, but whether the capabilities available to the project match what this particular study requires.

This task-based approach also prevents an obvious but surprisingly common mismatch: selecting a sophisticated methodology because it sounds appropriate, then discovering during data collection or analysis that nobody on the project can implement it properly.

Separate knowledge from practical competence

Understanding a method conceptually is not necessarily the same as being able to implement it competently.

You might understand what logistic regression does without being able to prepare the data, specify an appropriate model, assess relevant assumptions or diagnostics, interpret coefficients, and report the analysis correctly. You may understand thematic analysis from a textbook while having little experience developing codes, working systematically with qualitative material, documenting analytical decisions, or producing a defensible interpretation.

The distinction applies to technical procedures as well. Watching someone use laboratory equipment is different from being able to perform the procedure reliably yourself.

A study may require both. Do not infer practical readiness solely from having encountered a technique in a research-methods course.

Audit the entire research workflow

Skill gaps can occur at any stage, not only during statistical analysis.

Consider the complete path from research question to defensible conclusion. Can the study team formulate an appropriate design? Develop or select suitable measures? Implement sampling and recruitment procedures? Collect data consistently? Protect and organize research data? Conduct the planned analysis? Interpret the findings without exceeding what the design supports?

Research guidance from the University of Sheffield, for example, treats research methods, project planning, ethics, research-data management, and quantitative analysis as distinct areas in which researchers may need knowledge and support. That broader view is useful because methodological competence extends beyond knowing which statistical test to click in software.

Methodological design is itself a skill

A technically competent analyst cannot necessarily rescue a poorly designed study after data collection.

Researchers need enough methodological understanding to connect the research question with an appropriate design, sampling strategy, measurement approach, and analysis. This does not mean every investigator must be a professional methodologist. It does mean that methodological decisions should be made by people capable of recognizing the assumptions and consequences involved.

If you are uncertain whether the proposed design can answer the question, seek methodological input before the study is fixed. Consultation is considerably more useful when the design can still change than when someone is asked to repair an irreparable problem after the data have been collected.

Quantitative research may require more than knowing statistical software

Being able to operate SPSS, R, Stata, SAS, Python, or another analytical package does not by itself establish statistical competence.

Software executes instructions. The researcher still needs to understand why a particular analysis is appropriate, how variables should be represented, which assumptions matter, how the design affects analysis, what diagnostics or sensitivity analyses may be needed, and how results should be interpreted.

Statistical training commonly integrates these elements. For example, the University of the Philippines School of Statistics offers training that combines statistical concepts with sampling, inference, data management, exploratory analysis, programming, and appropriate software use rather than treating software operation as a substitute for statistical reasoning.

If your study requires an analysis you have never performed, determine what learning is needed beyond memorizing the sequence of commands.

Qualitative methods require methodological competence too

Qualitative research is sometimes treated as an easier alternative when researchers are uncomfortable with statistics. That is a poor reason to choose it.

Depending on the methodology, qualitative research may require competence in participant selection, interviewing or observation, reflexivity, data management, transcription decisions, coding, iterative analysis, interpretation, and methodological documentation. Different qualitative traditions may also make different assumptions about data generation and analysis.

Conducting a conversation is not automatically equivalent to conducting a research interview. Highlighting passages in transcripts is not automatically a defensible qualitative analysis.

The same standard should apply across methodologies: choose the approach because it fits the research question, then ensure that the project has the competence needed to implement it properly.

Data management can become a major skill requirement

Some studies are analytically straightforward but technically demanding because of the data.

You may need to merge several files, reshape longitudinal records, create reproducible transformations, manage identifiers, document derived variables, handle missing values, maintain versioned code, or work inside a secure research environment. Large datasets may require programming and computing skills beyond what ordinary spreadsheet software can reasonably support.

For qualitative projects, data-management requirements may include systematic organization of recordings, transcripts, coding files, field notes, and analytic memos while maintaining appropriate confidentiality protections.

Ask not only whether you can perform the final analysis but whether you can turn the raw evidence into a trustworthy analytical dataset or corpus.

Some studies require specialized technical or laboratory skills

Research may depend on capabilities that cannot be acquired from a weekend tutorial.

Laboratory assays, clinical measurements, imaging procedures, specialized equipment, advanced programming, geospatial analysis, psychometric modeling, machine learning, complex simulation, or other specialized procedures may require substantial training and supervised practice.

The consequences of poor execution also differ. An error in exploratory code may be reversible. Mishandling a specimen or performing an irreversible procedure incorrectly may destroy evidence that cannot simply be recollected.

Consider both the difficulty of learning the technique and the consequences of performing it badly.

Ethics and data protection can require practical competence

Researchers working with human participants or sensitive information need to understand the procedures relevant to their study, not merely obtain an approval letter.

This can include informed-consent procedures, privacy and confidentiality protections, secure data handling, management of participant identifiers, appropriate recruitment, handling of sensitive disclosures, and compliance with applicable institutional requirements.

The exact obligations vary across studies, institutions, and jurisdictions. If the project involves a procedure you do not understand, seek appropriate training or guidance before implementation.

Ethics is not a skill that can be outsourced entirely to the review committee. The research team has to conduct the approved study responsibly after approval is granted.

Do not forget communication and project-management skills

Some projects fail operationally rather than statistically.

Participant recruitment requires communication and follow-up. Multisite studies require coordination. Interviews require rapport and listening. Collaborative research requires clear division of responsibilities. Long projects require time management, documentation, and monitoring.

Research supervisors have identified both generic and research-specific capabilities as important to successful student projects, including communication, time management, literature searching, scientific writing, statistical aptitude, and the ability to work through ethics-review processes.

A brilliant analytical plan is of limited use if the project cannot recruit participants, maintain records, coordinate collaborators, or reach the analysis stage on time.

Classify each skill by how critical it is

Not every gap deserves the same response.

A useful distinction is whether a skill is essential to the validity or safe execution of the study, helpful for efficiency, or merely convenient.

This is more informative than labeling yourself simply "skilled" or "unskilled."

Ask whether the skill can realistically be learned in time

A new method is not automatically infeasible. The relevant issue is the learning curve relative to the project timeline.

Suppose you have eight months before analysis and need to learn basic data manipulation and regression in R. With appropriate training, practice, and supervision, that may be realistic. If the dataset arrives next week and your dissertation requires a highly specialized model you have never encountered, the situation is different.

Look for evidence of what training involves. Formal offerings can provide a useful reality check. Training in advanced quantitative methods may span weeks rather than hours and often assumes prior statistical knowledge. The Philippine Center for Economic Development and UP School of Economics, for example, have offered intensive impact-evaluation training covering econometrics, programming, experimental and quasi-experimental methods, and statistical software across several weeks.

The lesson is not that every researcher needs formal training. It is that learning time should be estimated rather than assumed away.

Practice on something that cannot damage the study

If a skill is new, your first attempt should ideally not be the only opportunity to produce the study's final evidence.

Practice interviews can reveal weak questions and interviewing habits. Pilot coding can expose ambiguities in a coding framework. A simulated or practice dataset can help you learn an analytical workflow before the real data arrive. Laboratory procedures may require supervised training and competency assessment before independent use.

Practice also provides better information about how long the task will take. A method that looked manageable in a textbook may turn out to require considerably more preparation in your specific context.

Supervision can close some gaps, but only if the support is real

Researchers sometimes list expertise as available because a supervisor, colleague, or statistician exists somewhere in the institution.

Availability needs to be more concrete.

Can the person actually advise your project? At what stages? How much time can they provide? Are they expected to teach you the method, review your work, perform part of the analysis, or become a collaborator? Will they still be available when the critical task occurs?

A specialist who might answer one question by email is different from a collaborator needed to design and conduct a complex analysis.

If your study depends on someone else's expertise, treat that expertise as a project resource that needs to be secured, not as ambient academic oxygen.

Some expertise should be involved before data collection

One of the costliest mistakes is waiting until analysis to discover that specialist input was needed during design.

Biostatistical collaboration provides a clear example. Guidance on effective investigator-biostatistician collaboration emphasizes involvement across study design, data collection, analysis, interpretation, and dissemination. NIH statistical collaboration procedures similarly describe statisticians working with investigators during early protocol development, including sample-size calculations and statistical analysis planning.

A statistician consulted only after an inadequate sample has been collected cannot retroactively increase the sample. A methodologist cannot reconstruct a comparison group that was never included. A data specialist cannot recover variables that were never measured.

If the specialist's expertise could change what data you collect or how the study is designed, involve that person before those decisions become irreversible.

Collaboration does not remove your responsibility to understand the study

Bringing in a specialist does not mean you can stop understanding that part of the research.

You should still understand the purpose of the method, the major assumptions relevant to the study, what information the specialist requires, and what the resulting analysis can and cannot establish. You also need enough understanding to interpret and communicate the findings responsibly.

Effective statistical consultation is often educational as well as technical. Research on biostatistical consultation has found that teaching and methodological explanation commonly occur during consultations, reflecting the fact that collaboration works best when investigators and specialists can communicate about both the scientific question and the analytical method.

The division of expertise may be unequal, but the research argument still needs to make sense as a whole.

Software competence and methodological competence are different

A researcher can know a statistical method and be unfamiliar with the software needed to implement it. The reverse is also possible.

Both may be necessary. Knowing where the regression button is located does not establish that regression is appropriate. Knowing regression theory does not guarantee that you can reliably prepare and analyze a complex dataset in unfamiliar software before Friday.

AI tools do not make a missing research skill disappear

Generative AI and other computational tools can assist with coding, explanations, data manipulation, drafting analytical syntax, and other research tasks. They may be useful learning and productivity aids where their use is appropriate and permitted.

They do not eliminate the need for competent oversight.

If you cannot evaluate whether generated code implements the intended analysis, whether an explanation is statistically correct, whether a qualitative coding suggestion fits the methodological framework, or whether a fabricated function or reference has appeared, the tool has not closed the competence gap. It may simply have made the gap harder to see.

Use tools to augment skills you can verify, not as evidence that expertise is no longer necessary.

Skill gaps have costs and timelines

Closing a gap may require a course, workshop, books, practice, software, specialist consultation, supervision, or repeated feedback. Those resources have costs in time and sometimes money.

A study that appears feasible when expertise is assumed to be free and immediately available may look different once training and specialist involvement are included.

Add those requirements to the project schedule. If you need six weeks to learn a technique before analysis, those six weeks belong in the estimate of whether the study can be completed within your deadline.

The goal is not to eliminate every learning challenge

A study designed entirely around skills you already possess may be unnecessarily restrictive. Research training is supposed to involve development.

The better distinction is between a productive stretch and a critical unsupported dependency.

A productive stretch asks you to learn a technique for which you have adequate foundations, time, resources, and supervision. A critical unsupported dependency asks the study to succeed only if you rapidly master something substantially beyond your preparation without reliable help.

The first can be excellent training. The second is a feasibility risk wearing a pedagogical hat.

Sometimes the study should be redesigned

If an essential skill cannot be learned or obtained within the available time, redesign may be more defensible than attempting the technique poorly.

This does not mean automatically replacing every advanced method with the simplest possible alternative. The revised method must still answer the research question.

You might narrow the question, use a less technically demanding but still appropriate design, choose data compatible with methods you can implement competently, reduce unnecessary methodological complexity, or bring in specialist expertise.

The adjacent question is therefore not simply whether the analysis is difficult, but whether you should avoid a research question because its analysis is beyond your current skills. Current ability and eventual project capability are not the same thing.

When an Ambitious Analysis Exceeds Your Current Skills

The student did not need to abandon the question merely because the method was unfamiliar. Nor would it have been sensible to assume that several online tutorials would automatically solve the problem. Feasibility came from identifying the gap, allowing enough learning time, securing relevant expertise, and involving that expertise before the design became fixed.

Common Mistakes When Judging Whether You Have the Required Research Skills

Create a Skills Map Before the Study Begins

Translate the methodology into a list of tasks and assign each one to the person who will actually perform or supervise it. Then classify the required competence as already available, learnable within the project, dependent on specialist support, or currently unsupported.

This turns an uncomfortable question about whether you are "qualified enough" into a practical resource assessment.

For particularly specialized requirements, the next question is when to bring in a statistician, methodologist, or other specialist. The best time is often earlier than researchers expect, especially when specialist input could change the study design rather than merely the final analysis.

Do You Have the Skills Your Study Requires?

Frequently Asked Questions About Research Skills and Feasibility

Your Study Needs the Required Expertise, but You Do Not Have to Possess All of It Today

How to Cite This Guide

This guide is intended to be read, shared, and used in research, teaching, and academic work. If you draw on its ideas, explanations, or other content, please acknowledge the source by citing the guide. Doing so gives appropriate credit and helps your readers locate the original resource.