Thing 12: How to Conduct Research Efficiently

Dr Simon Moss, Dean of Graduate Research and Emily Keough, Research Development Coordinator, University of Wollongong

Photo by Nick Morrison on Unsplash

Introduction

In general, people tend to underestimate the duration they need to complete tasks, called the planning fallacy. To illustrate, postgraduate researchers and other researchers often need more time to complete their research than anticipated. Their research is often delayed. Because of these delays

• researchers often experience considerable stress and anxiety,

• once they are ready to publish, the research questions may no longer be as relevant or significant.
  

How can you prevent these delays and progress on your research as efficiently as possible? To be consistent with this emphasis on efficiency, this Thing first outlines the three most useful and universal strategies to prevent delays—before exploring this topic in more depth. Specifically, this Thing first recommends you

• use generative AI as effectively and ethically as possible,

• learn how to confine your data collection and analysis without jeopardising the quality of your thesis,

• develop the capacity to write more fluently.

Recommendation 1: use generative AI effectively

Although many resources are available to help you utilise generative AI—such as Chat GPT or Google Gemini—as effectively and ethically as possible, this section considers the most overlooked features to save you time. First, learn how to ask prompts effectively. To illustrate,

• before you ask the question, inform the tool of your role, your university, and your research topic, as well as include a phrase like “Please adopt the role of a pedantic reviewer”,

• ask your question politely, because courtesy tends to improve responses.

• after you ask your question, include the phrase “Please commence your answer with the phrase ‘My best guess is…’—a strategy that limits misleading responses,

• after you receive an answer, include the prompt “Can you appraise and improve this response?”

You can even ask the AI tool to offer advice on how to improve your prompts.

Second, learn about some of the overlooked uses of AI tools. The following table illustrates some examples.

Practice	Examples
Data collection
Ask the tool to identify the range of methods that can measure or manipulate some variable, evaluate the cost, efficiency, and validity of these methods, help you administer or apply these methods effectively. For all your expenses, such as the costs of equipment, ask the tool to identify free alternatives.	I am studying whether age affects the relationship between exercise and diabetes. Which tools can I use to measure physical activity? Which tools can I use to measure the risk of diabetes? What is the cost of each measure in US dollars? Which measure is the quickest to administer? Which measure is the most valid or accurate? Can you suggest some papers I should read to support these claims? When applying these methods, what are the most common mistakes that people commit? How can I avoid these mistakes? I plan to purchase this equipment and these measures…Are you aware of free alternatives I could use instead?
Develop skills to conduct research proficiently
Use the tool to improve your capacity to implement some of these methods	I want to conduct qualitative research that examines the experience of international Higher Degree by Research (HDR) candidates who report a disability. I would like you to assume the role of an international HDR candidate with impaired vision. Can I interview you—so I can develop my interviewing skills? And can you provide feedback about how I could improve my questions. How could I have improved this interview? First, tell me about why you chose to study a PhD…
Data analysis
To analyse data efficiently, you could prompt the tool to not only suggest relevant techniques but also to clarify how to conduct these techniques, identify and overcome the challenges of this technique.	I asked 100 participants to specify their duration of exercise and their age. Half of these participants had been diagnosed with diabetes. Which statistical technique should I use to determine whether age affects the relationship between duration of exercise and incidence of diabetes. How can I conduct this technique? What software or tools should I use? What are some mistakes that researchers sometimes commit when they conduct this technique? How can I overcome or prevent these complications? What python or R code should I use to conduct this technique?
Discussion
Prompt the tool to determine the characteristics or conditions that affect some relationship between two or more variables. This information can help you identify and to explain limitations in the literature.	Here are 17 abstracts of studies that examined the association between exercise and diabetes. From these abstracts, can you identify the factors that moderate this association between exercise and diabetes.
Other
If you experience some challenge—such as a problem with proprietary software—prompt these tools to offer guidance.	How can I change the default language in MS Word on a Mac computer?
Learn about other AI tools	“Scholarly Assistant” can effectively summarise the latest research on a topic “Research Kick” can generate novel research questions. “Elicit” conducts systematic reviews.

Finally, you need first need to be attuned to some limitations that many universities impose—such as

• the warning to never upload copyrighted publications, private information, or intellectual property to insecure AI tools,

• to never use the answers verbatim, and

• to check all answers.

Recommendation 2: confine your research

Besides AI, the next main opportunity to conduct research more efficiently revolves around how to generate more conclusions from less data or fewer research activities as well as how to prevent delays. You can thus submit theses or publications after a shorter duration. The following table outlines some approaches you may consider to fulfill this goal.

People often feel their life in general, and their research in particular, are bathed with problems. The following table epitomises the problems that you may experience as you pursue your research.

Practice	Examples and Details
For every study, collect more data than needed—especially if these data can be collected easily and inexpensively.	For instance, if you administer surveys, include other questions or scales that comprise only a few items—questions or scales that do not appreciably increase the length of these surveys. If you need to conduct interviews, ask a couple of additional questions that are only peripherally related to your research question. Often, if your study does not generate the anticipated results, these other data may generate unforeseen insights that can be explored and validated in later studies. Similarly, pilot studies, feasibility studies, or other preliminary studies could include a few more participants, measures, or other activities. These preliminary studies could thus be converted into full studies.
Diminish the number of features or goals in your study.	For example, suppose you need to develop a prototype or intervention. Initially, you might develop a prototype that comprises fewer functions—or an intervention that comprises fewer activities—than initially planned. This prototype or intervention could then be assessed to determine whether the overlooked functions or activities are necessary. As this example shows, when you limit a prototype, intervention, or some other feature of your research, you do not only conserve time—but also explore whether a more confined approach is sufficiently effective. These limits, therefore, might uncover helpful insights.
Identify all the activities in your project that might be delayed or prolonged. Attempt to commence these activities as soon as possible.	For example, if you need to organise access to a database, attempt to arrange this access as soon as you can—and certainly before you are ready to analyse the data. Do not wait until you need these resources. In addition, attempt to organise alternatives to your plans even before these alternatives may be needed. So, for instance, suppose you might need to arrange another dataset in case you do not receive the first dataset you expect. Perhaps initiate this access to the second dataset from the outset.
Attempt to associate with people who network extensively or teams that collaborative extensively.	If you extend your networks, you might learn of other resources or opportunities you could utilise in response to impediments. For example, you might learn about another laboratory in which you can utilise equipment that has been delayed.
Consult the literature on rapid approaches—shorter variants of classical techniques that consume less time.	Indeed, scholars have developed a range of these rapid approaches, especially in qualitative research, including short-term ethnographies (Pink & Morgan, 2013) rapid appraisals (Beebe, 1995) or rapid qualitative inquiry—useful if two or more researchers are collaborating and need to collect and to analyse data within 6 weeks (Beebe, 2014) Rapid Assessment Procedures (Scrimshaw & Hurtado, 1988; Renfro et al., 2022) Rapid Ethnographic Assessments (Bentley et al., 1988) the RARE model (Brown et al., 2008), rapid reviews—comprising a series of techniques to synthesis the literature more efficiently (Tricco et al., 2018), and quick ethnography (Handwerker, 2001) You can also prompt generative AI tools to suggest other rapid approaches that might be useful. In general, rapid approaches may not be as accurate or informative as more comprehensive alternatives, but the time you save may enable you to complete more research, ultimately enhancing the quality of your work.

Recommendation 3: fluent writing

Perhaps the most vital skill to develop is your capacity to write precisely, concisely, professionally, but fluently. To illustrate the importance of this skill

• researchers who have not developed this skill often feel the need to rewrite many of their sentences,

• consequently, they can write only about 100 to 500 words a day,

• and their supervisors or collaborators may request several drafts,

• one paper might thus demand several months to write.
  

In contrast, some researchers learn many principles on how to write precisely, concisely, and professionally. The following table outlines some of these principles. Once researchers can apply these principles and similar insights,

• they no longer need to rewrite their sentences many times, because they write more effectively from the outset,

• if they first construct a plan of every argument they want to include in a paper, they can then usually write between 2000 and 3000 words a day,

• postgraduate researchers might save almost a year on their thesis.
  

The following table outline the most effective principles that researchers should learn.

key principles on how to write more proficiently and fluently

Avoid the word “it”—and replace this word with a more specific description. Sometimes phrases with the word “it” can be deleted.

Insert a noun or noun phrase after “this” or “these”. Similarly, replace ambiguous pronouns, like “those” and “others”, with more specific alternatives, such as “other people”.

Replace hazy verbs—such as came, come, do, done, for, get, got, give, gave, held, hold, look, view, seen, make, made, occur, put, run, ran, took, take, have, has, and had—with more precise alternatives. You can use a thesaurus to facilitate this task.

Replace “a number of” with a more specific quantity. Be as precise as possible. For example,

• “several phases” or “many phases” are better than “a number of phases”.

• but “5 phases” is better than “several phases” or “many phases”—but not always possible.

Specify the direction of relationships. That is, replace “impact”, “affect”, and “influence” with a positive or negative term. “This writing tutorial enhanced confidence” is more precise than “This writing tutorial affected confidence”.

Learn the difference between the following pairs of words if relevant to your research:

• effective vs efficient,

• equity vs equality,

• comprise vs include,

• fewer vs less,

• who vs that

• which vs that,

• affect vs effect,

• further vs farther.

Use consistent terms or names. Do not shift haphazardly to use a different term or name to refer to the same approach or concept.

To write concisely, condense phrase that include the words “aim to”, “aimed to”, “aims to”, “in terms of”, “in regard to”, “in respect to”, and “in relation to”. For example

• some of these words, such as “in terms of”, suggest the surrounding phrase can be omitted,

• some of these phrases, such as “in regard to”, can be reduced to one word, such as “about” or “on”.

To write concisely, attempt to diminish the length of verbose phrases—phrases that often include the words “of”, “there”, or “that”.

Here are some phrases that can often be reduced to one or two words, as shown in brackets

• at the present time (now)

• there is little doubt that (probably)

• despite the fact that (although)

• in the event that (if)

• for the purpose of (to)

• has an effect on (affects)

• on a daily basis (daily)

• during the course of (while)

Omit redundant words and phrases, such as “in order”, “it is noteworthy that”, “needless to say”, “in size”, “in shape”, “in colour”, “the month of”, and “a total of”.

Minimise the use of emotive words that are hard to justify, such as “very” and “extremely”. Use objective language rather than emotional or pejorative language.

Avoid terms that imply certainty. For example, indicate that a species has “not been detected” rather than “disappeared”.

Avoid contractions in formal writing, such as can’t, don’t, and won’t. Minimise the use of metaphors and informal or colloquial phrases.

In general, write short sentences. When researchers mainly write shorter sentences, the logic of their arguments evolves gradually and they are not as likely to commit grammatical errors. If you are a more advanced writer, you can occasionally write longer sentences.

Paragraphs should usually comprise between 3 and 8 sentences. Paragraphs that are too long often revolve around more than one specific argument. Instead, paragraphs should revolve around one specific argument—an argument that is usually outlined in one sentence, often the first.

So, how can you learn to write more proficiently and fluently. You could

• each day, apply one of the principles in the previous table to improve paragraphs you have already written,

• dedicate about 15 minutes a day to websites that help you develop this skill, such as this page,

• use generative AI tools to improve your writing skills.

Although you should never plagiarise the answers that AI tools generate, these tools can help you improve your writing capability. The following table illustrates how you can use these tools to achieve this goal.

Practice	Examples
Prompt the generative AI tool to help you identify more suitable terms.	In my research paper, I want to write about how I gave participants a survey. What is a better word or phrase to describe this activity?
Invite the generative AI tool to identify words that may be useful in your discipline.	Here is an excerpt of my writing. What are some helpful words that I have overlooked? In mechanical engineering, which words may be useful but perhaps overlooked by less experienced writers? What are some verbs or adjectives that are often used in science papers but seldom used in other literature?
Prompt the generative AI tool to help you write more precisely, concisely, and professionally.	Here is an excerpt of my writing. What are some principles I should consider to improve my writing? Can you illustrate these principles with reference to my paragraphs? In my research paper, I want to write “A lot of studies have looked at whether exercise really does help people with avoiding diabetes”. Which words could be more precise? Which words could be deleted or which phrases could be reduced?
Ask the generative AI tool to help you arrange your arguments in a logical order.	I want to describe evolutionary theory. What are the main propositions or assumptions of this theory? In what order should I present these propositions or assumptions?

Pod discussion

• What are some research activities that, in hindsight, you could have completed more efficiently? What lessons have you distilled from these experiences?

• Besides AI tools, what other technologies have you used to facilitate your research?
  

Simon Moss is the Dean of Graduate Research at the University of Wollongong. He is a registered psychologist and endorsed in organisational psychology. Since 1998, Simon has worked extensively in management consulting, especially in the fields of leadership, change management, collaboration, and conflict resolution. For example, he was also a cofounder of Zenith Professional Development, a company that collated every scientific discovery that contradicts prevailing management beliefs and practices.

Emily Keough is a Research Development Coordinator at the University of Wollongong. Her research revolves around experimental psychology, ranging from investigating the role of expression recognition ability in social functioning to exploring ways to reduce harmful weed and pathogen spread in Kosciusko National Park.

References

Beebe, J. (1995). Basic concepts and techniques of rapid appraisal. Human Organization, 54(1), 42–51.

Beebe, J. (2014). Rapid qualitative inquiry (2nd ed.). Rowman & Littlefield.

Bentley, M., Pelto, G., Straus, W., Schumann, D., Adegbola, C., de la Pena, E., Oni, G. A., Brown, K. H., & Huffman, S. L. (1988). Rapid ethnographic assessment: Applications in diarrhea management program. Social Science & Medicine, 27(1), 107-116.

Burgess-Allen, J., & Owen-Smith, V. (2010). Using mind mapping techniques for rapid qualitative data analysis in public participation processes. Health Expect, 13(4), 406–415.

Handwerker, P. (2001). Quick ethnography: A guide to rapid multi-method research. AltaMira Press.

Kruger, J., & Evans, M. (2004). If you don’t want to be late, enumerate: Unpacking reduces the planning fallacy. Journal of Experimental Social Psychology, 40, 586-598.

Pink, S., & Morgan, J. (2013). Short‐term ethnography: Intense routes to knowing. Symbolic Interaction, 36(3), 351-361.

Tricco, Andrea C., Langlois, Etienne. V., & Straus, Sharon E., Alliance for Health Policy and Systems Research, & World Health Organization. (2017). Rapid reviews to strengthen health policy and systems: A practical guide. World Health Organization.

Werner, C. M., Stoll, R., Birch, P., & White, P. H. (2002). Clinical validation and cognitive elaboration: Signs that encourage sustained recycling. Basic and Applied Social Psychology, 24, 185-203.