The Marketplace is now widely available! Install insights today.
Download AppBlogFeaturesPricingSupportSign In

Research methodology is a discipline of scientific procedures. It includes theory, analysis and guidelines for how research should proceed: how research should be conducted and the principles, procedures, and practices that direct research. Research methodology is the specific set of procedures or techniques used to identify, select, process, and analyse information about a topic. Since methodology can differ between different disciplines, therefore there is an assortment of different research methodologies which may not be appropriate for all research problems (Nayak and Singh 2015). Methodology should not be confused with scientific methods, which mean ways or techniques for gathering information/results. Scientific methods describe the way in which scientific knowledge is gained. In a research paper, the materials and methods section allows the reader to critically evaluate a study’s overall validity and reliability, because it states how the data were collected or generated, and how they were analysed. The following is an example of a research methodology:

  1. Observe and question: selection and definition of the research problem

  2. Review of the related literature

  3. Formulation of hypothesis

  4. Preparation of the research design, including sampling plan and selection of the tools for data collection

  5. Execution of the research plan: gathering the data

  6. Processing the data

  7. Report, including supporting or rejecting the hypothesis

Research designs

A research design is a blueprint for empirical research that includes planning, organizing and directing the research, including definition of the research problem, research questions, and objectives. It outlines how the research study will be carried out; therefore, it includes a thorough plan for data acquisition, definition of instruments used, and procedures for sampling and monitoring, in order to resolve specific research questions or to test a specific hypothesis. Research designs can be grouped into two categories:

  • Survey research design

  • Experimental research design

Survey research design

Surveys are mainly used in the social sciences. In surveys the data are collected from a pre-defined test group to gain information and understandings on various topics of interest. There are three different types of surveys according to their purpose: exploratory, descriptive, and explanatory studies (Nayak & Singh 2015).

Exploratory study or research usually starts with reviewing available data, or qualitative methods such as informal discussions, in-depth interviews, focus groups, and case studies; therefore, the data collected are qualitative. The data are then quantified and assumptions are drawn. Exploratory research cannot be generalized to the whole population. The results of the exploratory research cannot lead to firm conclusions, but they can enable important understanding of a given situation. The purpose of exploratory study is to frame a problem for a more exact investigation or to form hypotheses. Exploratory research studies therefore do not have hypotheses. Exploratory research design is used when little is known about the phenomenon and when earlier theories have failed to clarify it.

Descriptive study describes as exactly as possible the connection between the characteristics of a population and the studied phenomenon. It cannot describe what caused the situation, just what the characteristics are. Descriptive study is usually done after a survey and prior to explanatory study, so it is used when there is already some knowledge about a phenomenon, but we want to know more about it. Descriptive research studies therefore have hypotheses.

Explanatory study: When there is a known phenomenon that is sufficiently described, research proceeds by finding out the causes and reasons for it. The aim of explanatory research studies is to explain ‘why’. It goes beyond describing the problem and characteristics of the phenomenon, and aims to explain the causes and effects.

Experimental research design

Experimental research design is most common in environmental science. It is a real experiment, in which a researcher manipulates one variable and controls the other variables. Experimental research design provides evidence which contributes to a greater validity of the research. Experimental research always has a control group and a test group, in which a selected variable is manipulated (only one at a time), while extraneous variables are controlled. Experimental studies test a causal hypothesis, which refers to a causal relation between two variables where variable X (the cause) determines variable Y (the effect). Experimental studies thus aim to examine cause-effect relations (hypotheses) in strictly controlled conditions by separating the cause from the effect in time, exposing one group to the cause (the test or treatment group) while not exposing another group (the control group), and observing how the effects vary between these two groups. The main strong point of experimental design is the solid validity reached by isolation, control, and intensive examination of a small number of variables, while the main weakness is limited outward generalizability because situations in real life are frequently more complex and may include more extraneous variables than in artificial laboratory or field settings. Besides this, the researcher should identify all relevant extraneous variables and control them, otherwise the internal validity may be reduced, and false correlations may appear. The experiments can be carried out in a laboratory or in the field. Both ways have pros and cons. Laboratory experiments enable isolation of the target variables and control for the extraneous variables, which might not be the case in experiments in the field. Because of this, extrapolations made from laboratory experiments have a tendency to be stronger in internal validity, while those from field experiments tend to be stronger in external validity. Experimental data are processed by quantitative statistical methods (Nayak & Singh 2015).

Preliminary steps

Problem formulation

The first and most important step in research design is forming a problem. The problem should be identified and investigated. The problem cannot be explained successfully if a researcher does not have proper knowledge and understanding of specific issues causing or creating the problem. There are some main steps to follow when forming a problem (summarized by Nayak and Singh 2015):

  1. Define a research field

  2. The research field has to be well-known to the researcher carrying out the research (a specialist in the field)

  3. Review previous research conducted in the area in order to be familiar with recent findings

  4. Set the field of study, based on this review

  5. Identify the problem in general

  6. Identify the specific feature of the problem which is to be examined, and form a problem statement

A problem statement is an abstract of a problem formulation. It is important for further research design. Good problem statements focus on the relationship between two or more variables, are stated clearly and explicitly in a question form, can be tested empirically, and are not morally or ethically questionable.

Literature review

Following problem formulation, a systematic and detailed search of all types of scientific and expert literature referring to the research topic has to be executed with the aim of identifying a range of good quality references. The majority of the references should originate from peer-reviewed academic literature; however, other sources may also be relevant (legislation, publications of international organizations such as WHO and FAO, oral sources, etc.). The main academic literature consists of books, articles, journals, conference proceedings, research reports, databases, theses, and dissertations. After gathering all the information, a detailed review of the academic literature and a critical discussion of current knowledge have to be carried out. This is an important foundation for the success of the research project. The literature review gathers together the main theories and findings in the research area, identifies key authors, and highlights the gaps in knowledge that need to be focused on.

Nowadays literature review is mainly done with online searches in different databases. It is important to select appropriate keywords that can also be combined together using ‘and’ and ‘or’ to refine or specify the search results. Electronic journals and articles are the most up-to-date resources available. The papers can be published online as soon as they have been edited, with no need to wait for there to be enough papers to form the whole journal issue. This is of special importance in rapidly evolving fields (Nayak & Singh 2015). Some electronic resources are free (‘open access’). However, most have to be paid for. It is possible to purchase papers online as an individual researcher; however usually universities, libraries and other education institutions have paid subscriptions for different databases and their employees or members can access them for free. The most common academic databases and search engines are:

  • ScienceDirect is a leading full-text scientific database that includes journal articles from more than 2,500 journals and book chapters from almost 20,000 books

  • SpringerLink is the most comprehensive online collection of scientific, technological and medical journals, books and reference works

  • Google Scholar is a free search engine that catalogues academic information from various online web resources. It gathers information across a range of academic resources that are generally peer-reviewed. It is one of the most extensively used academic resources for researchers

  • Web of Science is a scientific citation indexing service for subscribed customers that provides a comprehensive citation search. It gives access to numerous databases

  • Mendeley research catalogue is a crowdsourced database of research documents. Researchers have uploaded nearly 100M documents into the catalogue with additional contributions coming directly from different repositories

  • PubMed is a database primarily of references and abstracts on life sciences and biomedical topics

  • Scopus is the world's largest abstract and citation database of peer-reviewed research literature. It contains over 20,500 titles from more than 5,000 international publishers. While it is a subscription product, authors can review and update their profiles via ORCID or by first searching for their profile at the free Scopus author lookup page

When preparing a literature review it is important to keep a database of the references, which can be used for making notes on the key points in each source (Nayak & Singh 2015). There are some software packages that enable the creation and organization of a personal database of scientific papers and the formation of citations when writing a scientific report. A database can be organized or browsed by authors, journals, date and other characteristics of the papers, or according to the topic, relevance, read/not read, favourites, etc. Particularly useful software packages for reference management are EndNote, Mendeley and RefWorks.

When a list of relevant articles is created, it is then necessary to peruse each article, or at least its abstract, to decide whether the article is suitable for a detailed review. Literature review should be comprehensive and not limited to a few papers, a few years, or a specific methodology. A literature review should examine if the primary research questions have previously been investigated, and what the outcomes were (in this case it should be explained why it is important to study them again), if there are novel or different research questions appearing, and if the primary research questions should be adapted or changed according to the findings from the literature. The literature review may also offer possible answers to the research questions, or help to identify theories that have formerly been used to discuss comparable questions (Nayak & Singh 2015).

A literature review is a well-structured and reasoned evaluative report of previous studies related to the research topic. The review provides description, evaluation and critique of this literature. It provides a theoretical foundation for the research and helps to determine its main characteristics. A literature review is more than gathering of information; it also comprises the identification of the relationship between the literature and the research topic.

Objectives of the study

In contrast to problem formulation, which describes the aim of the research, the objectives offer a definition of specific actions that will be taken to reach this aim. They describe what we expect to achieve by carrying out the research. There can be an overall objective followed by a list of specific objectives. The overall objective describes how we plan to address the problem: e.g. we need to find the answer to problem A by implementing action B. The specific objectives then describe action B in more detail. There are typically two to four specific objectives. Objectives thus explain how we are going to answer the research question. It is therefore a prerequisite that the research question is clear. Objectives usually begin with words such as: to identify, to establish, to describe, to determine, to estimate, to develop, to compare, to analyse, to collect, etc. (Nayak & Singh 2015). Good research objectives should be:

  • brief and precise

  • listed in a logical order as one objective may refer to another

  • realistic, meaning that it is possible to achieve them within the given timeframe and available resources

  • expressed in operative terms

  • unchanged from the beginning of the study (they should not be moving targets)


A hypothesis suggests a solution to the problem that is going to be empirically tested during the research and at the end it will be rejected or supported according to the observed results. The hypothesis is a guess or a proposal for generalisation (Nayak & Singh 2015). The hypothesis can be developed through analogy, induction, deduction, or intuition. The most important characteristic of a hypothesis is that it must be falsifiable, meaning that it can be disproven.

Hypotheses should be strong, not weak. An example of a weak hypothesis is ‘high concentrations of phosphorus are related to algae growth’, because it doesn’t indicate either the direction (i.e. if the relationship is positive or negative), nor the causality (i.e. if high concentrations of phosphorus cause algae growth, or if algae growth causes high phosphorus concentrations). A stronger hypothesis would be ‘high concentrations of phosphorus are positively related to algae growth’, which indicates the directionality but not the causality; and the strongest hypothesis would be ‘high phosphorus concentrations stimulate algae growth’, which postulates both the direction and the causality.

Protocol design

Protocol design is a written plan of the activities that have to be taken to sufficiently answer the stated research question. It includes choosing a research method for collecting data, and planning an appropriate sampling strategy to select a sample from the target population. The protocol should specify precisely:

  1. the characteristics of the test system (plant and fish species or varieties, source of supply, number, body weight span, illumination type and strength, etc.)

  2. detailed information on the experimental design, including a description of the chronological procedure of the study, all methods, materials and conditions, how many samples, what kind of samples, how many parallels, the dose levels and/or concentration(s), the type and frequency of analysis, measurements, observations and examinations to be performed, and the statistical methods to be used.

A sample is a smaller group of a population. The sample should represent the whole population in order to enable generalization of the outcomes from the research sample to the population as a whole. An appropriate sampling plan also provides a cost-effective use of research funds and appropriate research tempo, flexibility, and accuracy. There are two types of sampling (summarized by Nayak & Singh 2015): probability and non-probability sampling (Table 1). In probability sampling there is an equal chance for every subject or unit to be selected from the population, while in non- probability sampling all individuals in the population do not have an equal chance of being selected. This type of sampling is done when random sampling is impossible to do, when research has limited time, budget or workforce, or when the research does not aim at generalization to the whole population. In general non-probability sampling is more appropriate for the social sciences than for natural sciences. However, it can be used in a preliminary study to get some basic information about the population and to inform the kind of probability sampling to choose in an experiment. For example, we want to study the growth of lettuce in aquaponics, but we don't know if there are differences between the plants that grow on the edges of a raft and those that grow in the middle, so in a preliminary study we could take a few plants from the edge and a few from the middle (non-probability sampling), and measure them. If there are no differences between them, then simple random sampling can be used for the experiment, but if there are differences, then it would be better to use systematic random sampling or maybe even cluster sampling.

Table 1: Types of sampling

Type How the sample is gathered Additional explanation PROBABILITY SAMPLING Simple random By picking basic units in a way that each unit in the population has an equal chance of being picked A simple random sample does not have a sampling bias Systematic random By picking one unit randomly and picking additional units at uniform intervals until the required number of units is gained E.g. vegetables growing in a line and we pick every 5th vegetable Stratified random By independently picking an individual simple random sample from every single population stratum A population is divided into different strata (e.g. rafts or fish ponds) regarding specific characteristics or variables. The number of units we randomly pick from each stratum has to be aligned with the size of the stratum, since strata can be of different sizes; e.g. we may decide to pick 10% of units from each stratum Cluster A population is divided into clusters, and a sample is gathered by picking a few clusters by simple random sampling. The sample comprises a unit of randomly selected clusters The clusters are often made according to geographical/space units (e.g. all regions in a country; all rafts in aquaponics) while the analysis is done on randomly selected clusters (we randomly select the required number of whole rafts, which represents a sample) NON-PROBABILITY SAMPLING Convenience A sample is gathered from cases that are available for the study, i.e. ready to participate Purposive A sample is gathered from cases that have similar characteristics. The characteristics are selected in order to find answers to a specific question and can be most similar/dissimilar, most typical or critical. The prerequisite is that the researchers already know some characteristics of the population In contrast to stratified probability sampling where there is an equal chance to be selected for every unit in the same strata, in the case of purposive sampling the sample is non-randomly selected Snowball Is where existing study subjects recruit future subjects from among their acquaintances The sample group is said to grow like a rolling snowball. Also called chain sampling, chain-referral sampling, referral sampling Quota Divides the population into different groups similar to strata in stratified sampling (e.g. age, gender) A proportional or disproportional number of units is non-randomly chosen from every group

Besides selecting the appropriate type of sampling, sample size also has to be defined. The sample size depends on the characteristics of a population, mainly how heterogeneous it is. Besides this, the sample size is also related to the number of variables we want to analyse, the statistical procedures that we want to use, the desired precision, and the number of comparisons we want to make. On the other hand, sample size can also be limited by available time and funding.

There are several methods available to define the sample size to be used, including the Neyman- Pearson decision methodology or power analysis (Neyman & Pearson 1933). To estimate the sample size required we need an idea of the variance of the variable from the literature. The variance (and standard deviation) will depend on the variable considered and the species to be evaluated.

The data in natural sciences are mainly collected from observations and measurements using different laboratory and field instruments. Original records from the instruments and documentation, or their verified copies, which are the result of the original observations and activities, represent raw data. Raw data can be, for example, recorded data from automated instruments (e.g. O2, pH, EC probes), microscopic pictures, single measurements from laboratory instruments (e.g. readings from spectrophotometers), photographs, hand written observations (e.g. fish and plant health), and data from analogue measurements (e.g. analogue thermometer, settleable solids measured in Imhoff tank). Raw data must be converted into a computer-readable, numeric format, such as in a spreadsheet or a text file, so that they can be analysed by computer programs like R or SPSS Statistics.

A test system or unit of analysis is any biological, chemical, or physical system, or their combination, to be used in a study. It is a most basic element of research. The unit of analysis may be an organism or its part (e.g. fish), a colony or collective (e.g. vegetables), or an object (e.g. filtering system) that is the target of the investigation. The unit of analysis has to be defined at the beginning of a protocol design as it affects the instruments used and procedures taken during the research. Besides this, the lowest level of unit should always be chosen (e.g. collect data from separate plant tissues, not the whole plant together).

A test item is an item that is the subject of a study and a reference item (‘control item’) is an item used to provide a control for comparison with the test item.

A batch is a specific quantity, or a portion of test items or reference items formed by a defined experiment cycle in a way that it is expected that all items will have a uniform characteristic (e.g. one batch is lettuce under the same lighting conditions and different batches represent different lighting conditions).

Most project proposals include a section on the ethical aspects of the scientific protocols to be used. This usually involves previous approval of the methods by an ethics committee at the home institution, which mostly considers aspects related to animal welfare, in this case fish welfare. Those committees ask a set of questions including the justification for the research, its impact on the animals, and how distress can be prevented. For a set of guidelines about ethics, animal welfare and proper sampling procedures, see the NC3Rs Experimental Design Assistant, whose main aim is to replace, refine, and reduce the number of animals used in experimentation. It is thought that in the near future, scientists will be able to get their procedures and protocols approved by target journals before publishing the results, and thus have a certain guarantee that their studies will be published. This movement is called pre-registration (Nosek et al. 2018), and is aimed at strengthening methodologies and scientific results across the board. Finally, many journals are now asking that the raw data and results of published studies be made available in online databases, for example using the Data Research Item on Research Gate.

Good Laboratory Practice (GLP) means a quality system referring to the organisational process and the conditions under which studies are planned, performed, monitored, recorded, archived and reported (OECD 1998).

Standard Operating Procedures (SOPs) are documented procedures which describe how to perform tests or activities normally not specified in detail in study plans or test guidelines. SOPs include:

  1. maintaining the records including test and reference items characterisation, date of receipt, expiry date, quantities received and used in studies

  2. identification of handling, sampling, and storage procedures in order that homogeneity and stability are guaranteed as far as possible and contamination is avoided

  3. storage containers should be marked with identification information, expiry date, and specific storage instructions.

After deciding which subject to study, what to measure, and how to gather and analyse data, it is time to execute the research. Research execution also includes preliminary tests of the equipment, laboratory instruments, sampling and analyses. Preliminary testing is an important part of the research process since it enables the detection of potential problems in the research design and for the laboratory instruments used in the study to be checked so that they are reliable and provide valid measures. After preliminary testing the research design may be optimized and then the real research can be executed.

All data generated during the research should be recorded directly, promptly, accurately, and legibly in the laboratory diary. These entries should be signed and dated. In order to ensure traceability, a research project needs to have a unique identification, and all samples, specimens, data files etc. concerning the study should carry this same identification. Any change in the raw data should be made in a way that does not to delete the previous entry, the reason for any changes should be indicated, and the change needs to be dated, and signed by the individual who made it.

Analysis of results

Tables and figures

Tables and figures are the quickest way to communicate large amounts of complex information. They have to be designed carefully. A good table or figure should present the data simply, clearly, and neatly, and allow the reader to understand the results without having to look at other sections of the paper; i.e. tables and figures should be self-explanatory and understandable even when they are taken out of the text; therefore, clear and informative titles are crucial. A good figure (graph or picture) should have:

  • only the necessary information

  • large enough lettering

  • a frame

  • a legend that explains everything necessary

  • a graphical format in a high resolution (>300 dpi)

A good table should have:

  • ##### a separate cell for each value

  • only horizontal line borders

  • values with a reasonable number of digits after a decimal point Larger tables are published in supplements to scientific papers.

In order to report the results, valid and internationally recognized units of measurements must be used. In science, industry and medicine the International System of Units (abbreviated SI) is used. In some geographic locations (e.g. United States) the imperial system is used, which includes units such as gallons, feet, miles, pounds and ppm. This system is not appropriate for international scientific publications. The SI system includes seven base units (Table 1).

Table 1: Seven base units of the International System of Units

QuantityUnitSymbolMasskilogramkgTimesecondsTemperaturekelvinKElectric currentampereAThe amount of a substancemolemolLuminous intensitycandelacdDistancemeterm

The most important methodological choice researchers make is based on the distinction between qualitative and quantitative data. Qualitative data take the form of descriptions based on language or images, while quantitative data take the form of numbers. The choice of which methodology to use will depend on your research questions, the formulation of which is consequently informed by your research perspective. Social science research can generate both qualitative and quantitative data, typically through the use of surveys. The data are collected from a pre-defined test group in order to gain information and understanding on various topics of interest. There are various different types of survey methods, including questionnaires, informal discussions, in-depth interviews, focus groups, and case studies.

Qualitative data are richer and are generally grounded in a subjective perspective. However, while this is generally the case, it is not always so. Qualitative research supports an in-depth understanding of the situation investigated and, due to time constraints, it generally involves a small sample of participants. For this reason, the findings are limited to the sample studied and cannot be generalised to other contexts or to the wider population. Popular methods for generating qualitative data include semi-structured or unstructured interviews, participant observations, and document analysis. Good qualitative analysis is generally more time-consuming than quantitative analysis.

Quantitative data, on the other hand, might be easier to collect and analyse, and they are based on a large sample. Quantitative measurements involve collecting data that can be ‘objectively’ measured with numbers. The data are analysed through numerical comparisons and statistical analysis. For this reason, it appears more ‘scientific’ and may appeal to people who seek clear answers to specific causal questions. Quantitative analysis is often quicker to carry out as it involves the use of measuring equipment and software. Owing to the large number of samples it allows generalisation to a wider group than the research sample.

Experimental research, on the other hand, is most common in environmental science. In experiments, a researcher manipulates one variable and controls the other variables in order to explore cause-and-effect relations. The data collected are quantitative and can be analysed using appropriate statistical methods.

Research report publication

An experiment is not completed until the results have been published and understood. The publication of results is important to allow for the reproducibility of experiments; therefore, the methods are shown separately from the results. As stated by the Council of Biology Editors (1968) ‘an acceptable primary scientific publication must be the first disclosure of a research containing sufficient information to enable peers (1) to assess observations, (2) to repeat experiments, and (3) to evaluate the intellectual processes; moreover, it must be attractively formatted and transparent, essentially permanent, available to the scientific community without restriction, and available for regular screening by one or more of the major recognized secondary services‘ (e.g. Biological Abstracts, Chemical Abstracts) (CBE 1968).

Good scientific writing is simple writing. Science is complex, but the writing used to describe it does not need to be. The best writing is that which gives the sense in the fewest simple words. High- quality, simple writing:

  • increases the chances of acceptance for publication

  • increases the impact of a publication in the research community

  • accelerates the understanding and acceptance of research

  • increases the faith of the readers in the quality of the research

Poorly written and complicated manuscripts annoy readers, peer reviewers, and journal editors, and hinder their understanding of complicated scientific concepts. A submission is more likely to be accepted if it:

  • describes research that advances the field

  • is carefully prepared and formatted

  • uses clear and concise language

  • follows ethical standards

The publication process:

  1. A need/wish to publish

  2. Choose a journal according to: topics of the journal, the journal’s audience, types of articles, reputation of the journal, impact factor, or personal requirements. We can find appropriate journals by checking where similar papers have been published and by online searches

  3. Read back issues

  4. Write the first draft

  5. Use a critical friend for the first check

  6. Refine further drafts

  7. Check that the article adheres to the author guidelines

  8. Proofread and submit

There can be more than one author of a scientific publication. The co-authors are the people who made substantial intellectual contributions to a study that is going to be published. It is important to keep the number of co-authors at a reasonable amount: the first author is usually the one that led the research and did the majority of the writing, and the last author is usually the one who is the head of the research group. In between it is customary to put the co-authors in alphabetical order by their surname, e.g. Wilson, T., Abercombie, J., Brown, E., Curwen, H., Davenport, K. & Albert, W.

Scientific manuscripts are peer-reviewed manuscripts in journals and books that typically have an impact factor (IF). The IF is used to compare different journals within a certain field. Reports, conference papers, posters and talks are not scientific manuscripts and do not have an IF. IF is a measure reflecting the yearly average number of citations of articles in that journal. For journals listed in Journal Citation Reports, IFs are calculated yearly for the year before, following the formula below:



𝐼𝐹𝑦 = Impact Factor in year y

𝐶𝑖𝑡𝑎𝑡𝑖𝑜𝑛𝑠 = number of citations

𝑃𝑢𝑏𝑙𝑖𝑐𝑎𝑡𝑖𝑜𝑛𝑠 = number of articles published

𝑦 − 1 = current year minus one

𝑦 − 2 = current year minus two

All scientific articles follow the same prescribed structure. This structure provides a logical line through the contents, enables manuscripts to be predictable and easy to read, presents a ‘map’ so that the readers can quickly find the contents of interest in any manuscript and, last but not least, reminds authors what contents needs to be included. The structure is as follows:

  • Title

  • Abstract

  • ##### Introduction

  • Materials and Methods

  • ##### Results

  • Discussion

  • Conclusion

  • Acknowledgements

  • References

Besides stated chapters, each manuscript usually also includes table(s) and figures, and supplemental data in a separate file(s). The main contents of the scientific paper are described in the core chapters: Introduction (which problem we are going to study), Materials and Methods (how we are going to study the problem), Results (what we found out), and Discussion (what it means). According to the capital letters of the chapters, this structure is called IMRaD format.

Title and abstract

Title and abstract are the most visible parts of the article. They can be seen on the journal website and in databases (e.g. Science Direct, PubMed, etc.); therefore, it is important to pay appropriate attention to their formulation. A well-prepared abstract enables readers to identify the basic contents of a document quickly and accurately, to determine its relevance to their interests, and thus decide whether they need to read the document in its entirety (Day 1998).

The title must be as accurate, informative, and as complete as possible. It gives the first information to the reader who then decides whether to continue reading or not. It is therefore crucial that the title is as descriptive as possible. To achieve this, specific rather than general terms should be used; however, the title should still be understandable and reasonably simple. The title usually does not include abbreviations, acronyms, or initials. Any scientific names should be written in full (e.g. Lactuca sativa, rather than L. sativa).

The abstract usually contains 200-300 words. It must outline the most important aspects of the study: it has to include the background, methodology, and results, but in limited details. It should only reproduce the facts covered in the manuscript. It is advisable to include synonyms for words and concepts that are in the title and, as for scientific writing per se, an understandable and reasonably simple writing style should be used. On the other hand, the abstract should not include abbreviations or cite references.


The introduction should provide the information needed to understand the study, and the reasons why the experiments were conducted. It should explain what question/problem was studied, and give information from previous studies; therefore, it includes numerous citations. The latter should be well balanced, current, and relevant. The introduction is not a literature review, but literature reviews can be cited (Nayak & Singh 2015).

Materials and methods

Materials and methods provide all the details of how the study was conducted. Different methodologies that were used in the study can be divided by subheadings. Any new methods that were used should be described in enough detail such that another researcher can reproduce the experiment. Previously used and published methods should be cited, and any modification done to the established methods should be described accurately. All statistical tests and parameters should be listed. The materials and methods chapter should be written in the past tense.


The results chapter gives an overview of the experiments, without repeating the details, which were described in the methods. Besides this, the researcher should critically review the data and select the results that are going to be published. A simple transfer of the data from the lab diary into the manuscript will not suffice for efficient presentation of the results. The presentation should be transparent and representative and can be done through either text or tables and figures. The data already described in the tables or figures should not be described again in detail in the text. The tables and figures should be quoted in the text only briefly. If there is only one or a few measurements of a characteristic, then it is usually described in the text, while if it is repeated measurements, then a table or graph is more representative. Depending on the journal, results can form an individual chapter or be joined with the discussion into a single chapter. The results should be written in a logical order, and divided into subsections with short, informative headings. Results of statistical analyses should also be included and presented in the text. The results chapter should be written in the past tense, while the present tense is used for referring to tables and figures.


The majority of the discussion and conclusions chapter should be an interpretation of the results. Subchapters can be formed following the logical framework of the subchapters in the results chapter. In the discussion chapter, the results of the research are compared with previous studies. The limitations of the research also have to be described, any inconclusive results should be mentioned and, if the findings are preliminary, suggestions for future studies should be pointed out. The main conclusions should be repeated at the end of the discussion, or in a separate conclusions chapter.


When writing a scientific manuscript, it should always be clear what are the thoughts, evaluations, and text of the authors of this study, and what has been derived from the authors of other publications. The source should be provided for any statement that does not come from the writers of the manuscript, by writing the author and year of publication – for example, the microelement nickel plays an important role in the decomposition of urea in aquaponic systems (Komives & Junge 2018), while the complete citation is given in the references – for example, Komives, T. & Junge, R. 2018. Importance of nickel as a nutrient in aquaponic systems – some theoretical considerations. Ecocycles 4 (2), 1-3. The references should be written in a style as demanded by the journal where the manuscript is going to be published, and therefore the journal citation style in the Instructions for Authors has to be carefully checked. There are various software programmes that enable appropriate management of references (EndNote, Zotero, RefWorks, Mendeley etc) (see


Plagiarism is cheating and is morally wrong. It is the use of someone else’s work without acknowledgment, as if it were your own. To avoid it, one has to know how to document the use of other people’s work. A researcher is responsible for referencing the use of sources in every paper that he/she writes. There are two ways to reference the works of other authors:

  1. Paraphrasing means summarizing another author’s ideas in your own words, while still referring to the original source. Quotation marks are not required. A well-paraphrased statement is concise and demonstrates a researcher’s understanding of what he/she has read. When paraphrasing or referring to an idea from another publication, it is beneficial to provide a page or paragraph number for the reference, especially when citing a long and complex text (e.g. a book).

  2. Direct quotes mean a direct repetition of a statement and are rarely used in scientific writing. Quotations should be used economically, mainly for historical or political quotes from eminent persons. Quotations of the findings from previous researches have to be avoided as the reader also wants to see the writers’ views and analysis of what has been read, which is not given in the direct quote. When using a direct quote, it is necessary to put quotation marks at the beginning and at the end of the quote.

Copyright © Partners of the [email protected] Project. [email protected] is an Erasmus+ Strategic Partnership in Higher Education (2017-2020) led by the University of Greenwich, in collaboration with the Zurich University of Applied Sciences (Switzerland), the Technical University of Madrid (Spain), the University of Ljubljana and the Biotechnical Centre Naklo (Slovenia).

Please see the table of contents for more topics.

[email protected]

Stay up-to-date on the latest Aquaponic Tech


  • Our Team
  • Community
  • Press
  • Blog
  • Referral Program
  • Privacy Policy
  • Terms of Service

Copyright © 2019 Aquaponics AI. All rights reserved.