How did life begin? Why do we sleep? Why does time always run forward and never reverse? These are among the age-old questions that up to now, no definitive answers had been found.
In research, these are called empirical questions, which ask about how the world is, how the world does work, etc. Such questions are addressed by a corresponding type of study—called empirical research—which is concerned about actual events and phenomena.
Empirical research typically seeks to find a general story or explanation, one that applies to various cases and across time. It functions to create new knowledge about the way the world actually works. This article discusses the concepts, types, processes, and other important aspects of empirical research.
Empirical research is defined as any study whose conclusions are exclusively derived from concrete, verifiable evidence. The term empirical basically means that it is guided by scientific experimentation and/or evidence. Likewise, a study is empirical when it uses real-world evidence in investigating its assertions.
This research type is founded on the view that direct observation of phenomena is a proper way to measure reality and generate truth about the world (Bhattacharya, 2008). And by its name, it is a research approach that observes the rules of empiricism and uses quantitative and qualitative methods for gathering evidence.
For instance, a study is being conducted to determine if working from home helps in reducing stress from highly-demanding jobs. An experiment is conducted using two groups of employees, one working at their homes, the other working at the office. Each group was observed. The outcomes derived from this research will provide empirical evidence if working from home does help reduce stress or not.
It was the ancient Greek medical practitioners who originated the term empirical (empeirikos which means “experienced”) when they began to deviate from the long-observed dogmatic principles to start depending on observed phenomena. Later on, empiricism pertained to a theory of knowledge in philosophy, which follows the belief that knowledge comes from evidence and experience derived particularly using the senses.
What ancient philosophers considered as empirical research pertained to the reliance on observable data to design and test theories and reach conclusions. As such, empirical research is used to produce knowledge that is based on experience. At present, the word “empirical” pertains to the gathering of data using evidence that is derived through experience or observation or by using calibrated scientific tools.
Most of today’s outstanding empirical research outputs are published in prestigious journals. These scientific publications are considered high-impact journals because they publish research articles that tend to be the most cited in their fields.
View in full screen
Download PNG image
Download JPEG image
View data table
General Scientific Journals with Highest Impact Factors (2018)
General Scientific Journals with Highest Impact Factors (2018) Nature (since 1869): 43.070
Nature (since 1869)
General Scientific Journals with Highest Impact Factors (2018) Science (since 1880): 41.063
Science (since 1880)
General Scientific Journals with Highest Impact Factors (2018) Science Advances (since 2015): 12.804
Science Advances (since 2015)
General Scientific Journals with Highest Impact Factors (2018) Nature Communications (since 2010): 11.878
Nature Communications (since 2010)
General Scientific Journals with Highest Impact Factors (2018) Proceedings of the National Academy of Sciences (since 1914): 9.580
Proceedings of the National Academy of Sciences (since 1914)
General Scientific Journals with Highest Impact Factors (2018) Scientific Reports (since 2011): 4.011
Scientific Reports (since 2011)
General Scientific Journals with Highest Impact Factors (2018) Philosophical Transactions of the Royal Science A (since 1905, 1665 before split): 3.093
Philosophical Transactions of the Royal Science A (since 1905, 1665 before split)
General Scientific Journals with Highest Impact Factors (2018) Proceedings of the Royal Society A (since 1905, 1800 before split): 2.818
Proceedings of the Royal Society A (since 1905, 1800 before split)
General Scientific Journals with Highest Impact Factors (2018) PLOS ONE (since 2006): 2.766
PLOS ONE (since 2006)
Source: Clarivate Web of Science Journal Citations Report
II. Types and Methodologies of Empirical Research
Empirical research is done using either qualitative or quantitative methods.
Qualitative research – Qualitative research methods are utilized for gathering non-numerical data. It is used to determine the underlying reasons, views, or meanings from study participants or subjects. Under the qualitative research design, empirical studies had evolved to test the conventional concepts of evidence and truth while still observing the fundamental principles of recognizing the subjects beings studied as empirical (Powner, 2015).
This method can be semi-structured or unstructured. Results from this research type are more descriptive than predictive. It allows the researcher to draw a conclusion to support the hypothesis or theory being examined.
Due to realities like time and resources, the sample size of qualitative research is typically small. It is designed to offer in-depth information or more insight regarding the problem. Some of the most popular forms of methods are interviews, experiments, and focus groups.
Quantitative research – Quantitative research methods are used for gathering information via numerical data. This type is used to measure behavior, personal views, preferences, and other variables. Quantitative studies are in a more structured format, while the variables used are predetermined.
Data gathered from quantitative studies is analyzed to address the empirical questions. Some of the commonly used quantitative methods are polls, surveys, and longitudinal or cohort studies.
There are situations when using a single research method is not enough to adequately answer the questions being studied. In such cases, a combination of both qualitative and quantitative methods is necessary.
III. Qualitative Empirical Research Methods
Some research questions need to be gathered and analyzed qualitatively or quantitatively, depending on the nature of the study. Here are the general types of qualitative research methods.
This involves observing and gathering data from study subjects. As a qualitative approach, observation is quite personal and time-intensive. It is often used in ethnographic studies to obtain empirical evidence.
The observational method is a part of the ethnographic research design, e.g., archival research, survey, etc. However, while it is commonly used for qualitative purposes, observation is also utilized for quantitative research, such as when observing measurable variables like weight, age, scale, etc.
One remarkable observational research was conducted by Abbott et al. (2016), a team of physicists from the Advanced Laser Interferometer Gravitational-Wave Observatory who examined the very first direct observation of gravitational waves. According to Google Scholar’s (2019) Metrics ranking, this study is among the most highly cited articles from the world’s most influential journals (Crew, 2019).
This method is exclusively qualitative and is one of the most widely used (Jamshed, 2014). Its popularity is mainly due to its ability to allow researchers to obtain precise, relevant information if the correct questions are asked.
This method is a form of a conversational approach, where in-depth data can be obtained. Interviews are commonly used in the social sciences and humanities, such as for interviewing resource persons.
This method is used to identify extensive information through an in-depth analysis of existing cases. It is typically used to obtain empirical evidence for investigating problems or business studies.
When conducting case studies, the researcher must carefully perform the analysis, ensuring the variables and parameters in the current case are similar to the case being examined. From the findings of a case study, conclusions can be deduced about the topic being investigated.
Case studies are commonly used in studying the experience of organizations, groups of persons, geographic locations, etc.
This primarily involves the process of describing, interpreting, and understanding textual content. It typically seeks to connect the text to a broader artistic, cultural, political, or social context (Fairclough, 2003).
A relatively new research method, textual analysis is often used nowadays to elaborate on the trends and patterns of media content, especially social media. Data obtained from this approach are primarily used to determine customer buying habits and preferences for product development, and designing marketing campaigns.
A focus group is a thoroughly planned discussion guided by a moderator and conducted to derive opinions on a designated topic. Essentially a group interview or collective conversation, this method offers a notably meaningful approach to think through particular issues or concerns (Kamberelis & Dimitriadis, 2011).
This research method is used when a researcher wants to know the answers to “how,” “what,” and “why” questions. Nowadays, focus groups are among the most widely used methods by consumer product producers for designing and/or improving products that people prefer.
View in full screen
Download PNG image
Download JPEG image
View data table
Google Scholar's Most Highly Cited Articles from the Most Influential Journals 2019
Google Scholar's Most Highly Cited Articles from the Most Influential Journals 2019 Deep Residual Learning for Image Recognition (He et al., 2016): 25256
Deep Residual Learning for Image Recognition (He et al., 2016)
Google Scholar's Most Highly Cited Articles from the Most Influential Journals 2019 Deep Learning (Lecun et al., 2015): 16750
Deep Learning (Lecun et al., 2015)
Google Scholar's Most Highly Cited Articles from the Most Influential Journals 2019 Going Deeper with Convolutions (Szegedy et al., 2015): 14424
Going Deeper with Convolutions (Szegedy et al., 2015)
Google Scholar's Most Highly Cited Articles from the Most Influential Journals 2019 Fully Convolutional Networks for Semantic Segmentation (Long et al., 2015): 10153
Fully Convolutional Networks for Semantic Segmentation (Long et al., 2015)
Google Scholar's Most Highly Cited Articles from the Most Influential Journals 2019 Prevalence of Childhood and Adult Obesity in the United States (Ogden et al., 2014): 8057
Prevalence of Childhood and Adult Obesity in the United States (Ogden et al., 2014)
Google Scholar's Most Highly Cited Articles from the Most Influential Journals 2019 Global, regional and national prevalence of overweight and obesity (Ng et al., 2014): 7371
Global, regional and national prevalence of overweight and obesity (Ng et al., 2014)
Google Scholar's Most Highly Cited Articles from the Most Influential Journals 2019 Observation of Gravitational Waves from a Binary Black Hole Merger (Abbott et al., 2016): 6009
Observation of Gravitational Waves from a Binary Black Hole Merger (Abbott et al., 2016)
Source: Google Scholar (2019); Crew (2019).
IV. Quantitative Empirical Research Methods
Quantitative methods primarily help researchers to better analyze the gathered evidence. Here are the most common types of quantitative research techniques:
A research hypothesis is commonly tested using an experiment, which involves the creation of a controlled environment where the variables are maneuvered. Aside from determining the cause and effect, this method helps in knowing testing outcomes, such as when altering or removing variables.
Traditionally, experimental, laboratory-based research is used to advance knowledge in the physical and life sciences, including psychology. In recent decades, more and more social scientists are also adopting lab experiments (Falk & Heckman, 2009).
Survey research is designed to generate statistical data about a target audience (Fowler, 2014). Surveys can involve large, medium, or small populations and can either be a one-time event or a continuing process
Governments across the world are among the heavy users of continuing surveys, such as for census of populations or labor force surveys. This is a quantitative method that uses predetermined sets of closed questions that are easy to answer, thus enabling the gathering and analysis of large data sets.
In the past, surveys used to be expensive and time-consuming. But with the advancement in technology, new survey tools like social media and emails have made this research method easier and cheaper.
This method leverages the strength of comparison. It is primarily utilized to determine the cause and effect relationship among variables (Schenker & Rumrill, 2004).
For instance, a causal-comparative study measured the productivity of employees in an organization that allows remote work setup and compared that to the staff of another organization that does not offer work from home arrangements.
While the observation method considers study subjects at a given point in time, cross-sectional research focuses on the similarity in all variables except the one being studied.
This type does not allow for the determination of cause-effect relationships since subjects are now observed continuously. A cross-sectional study is often followed by longitudinal research to determine the precise causes. It is used mainly by pharmaceutical firms and retailers.
A longitudinal method of research is used for understanding the traits or behavior of a subject under observation after repeatedly testing the subject over a certain period of time. Data collected using this method can be qualitative or quantitative in nature.
A commonly-used form of longitudinal research is the cohort study. For instance, in 1951, a cohort study called the British Doctors Study (Doll et al., 2004) was initiated, which compared smokers and non-smokers in the UK. The study continued through 2001. As early as 1956, the study gave undeniable proof of the direct link between smoking and the incidence of lung cancer.
This method is used to determine the relationships and prevalence among variables (Curtis et al., 2016). It commonly employs regression as the statistical treatment for predicting the study’s outcomes, which can only be a negative, neutral, or positive correlation.
A classic example of this research is when studying if high education helps in obtaining better-paying jobs. If outcomes indicate that higher education does allow individuals to have high-salaried jobs, then it follows that people with less education tend to have lower-paying jobs.
V. Steps for Conducting Empirical Research
Since empirical research is based on observation and capturing experiences, it is important to plan the steps to conduct the experiment and how to analyze it. This will enable the researcher to resolve problems or obstacles, which can occur during the experiment.
Step #1: Establishing the research objective
In this initial step, the researcher must be clear about what he or she precisely wants to do in the study. He or she should likewise frame the problem statement, plans of action, and determine any potential issues with the available resources, schedule, etc. for the research.
Most importantly, the researcher must be able to ascertain whether the study will be more beneficial than the cost it will incur.
Step #2: Reviewing relevant literature and supporting theories
The researcher must determine relevant theories or models to his or her research problem. If there are any such theories or models, they must understand how it can help in supporting the study outcomes.
Relevant literature must also be consulted. The researcher must be able to identify previous studies that examined similar problems or subjects, as well as determine the issues encountered.
Step #3: Framing the hypothesis and measurement
The researcher must frame an initial hypothesis or educated guess that could be the likely outcome. Variables must be established, along with the research context.
Units of measurements should also be defined, including the allowable margin of errors. The researcher must determine if the selected measures will be accepted by other scholars.
Step #4: Defining the research design, methodology, and data collection techniques
Before proceeding with the study, the researcher must establish an appropriate approach for the research. He or she must organize experiments to gather data that will allow him or her to frame the hypothesis.
The researcher should also decide whether he or she will use a nonexperimental or experimental technique to perform the study. Likewise, the type of research design will depend on the type of study being conducted.
Finally, the researcher must determine the parameters that will influence the validity of the research design. Data gathering must be performed by selecting suitable samples based on the research question. After gathering the empirical data, the analysis follows.
Step #5: Conducting data analysis and framing the results
Data analysis is done either quantitatively or qualitatively. Depending on the nature of the study, the researcher must determine which method of data analysis is the appropriate one, or whether a combination of the two is suitable.
The outcomes of this step determine if the hypothesis is supported or rejected. This is why data analysis is considered as one of the most crucial steps in any research undertaking.
Step #6: Making conclusions
A report must be prepared in that it presents the findings and the entire research proceeding. If the researcher intends to disseminate his or her findings to a wider audience, the report will be converted into an article for publication.
Aside from including the typical parts from the introduction and literature view, up to the methods, analysis, and conclusions, the researcher should also make recommendations for further research on his or her topic.
View in full screen
Download PNG image
Download JPEG image
View data table
Empirical researches with the largest number of study subjects (so far)
Empirical researches with the largest number of study subjects (so far) Twitter users (Fake News Study, Vosoughi et al., 2018): 3000000
Twitter users (Fake News Study, Vosoughi et al., 2018)
Empirical researches with the largest number of study subjects (so far) Adults (Autism study, Greenberg et al., 2018): 671606
Adults (Autism study, Greenberg et al., 2018)
Empirical researches with the largest number of study subjects (so far) Danish children (Vaccines & Autism, Hviid et al., 2019): 657461
Danish children (Vaccines & Autism, Hviid et al., 2019)
Empirical researches with the largest number of study subjects (so far) British Adults (Sleep duration study, Dashti et al., 2019): 446000
British Adults (Sleep duration study, Dashti et al., 2019)
Sources: Various Researches (cited above)
VI. Empirical Research Cycle
The empirical research cycle is composed of five phases, with each one considered as important as the next phase (de Groot, 1969). This rigorous and systematic method can consistently capture the process of framing hypotheses on how certain subjects behave or function and then testing them versus empirical data. It is considered to typify the deductive approach to science.
These are the five phases of the empirical research cycle:
During this initial phase, an idea is triggered for presenting a hypothesis. It involves the use of observation to gather empirical data. For example:
Consumers tend to consult first their smartphones before buying something in-store.
Inductive reasoning is then conducted to frame a general conclusion from the data gathered through observation. For example:
As mentioned earlier, most consumers tend to consult first their smartphones before buying something in-store.
A researcher may pose the question, “Does the tendency to use a smartphone indicate that today’s consumers need to be informed before making purchasing decisions?” The researcher can assume that is the case. Nonetheless, since it is still just a supposition, an experiment must be conducted to support or reject this hypothesis.
The researcher decided to conduct an online survey to inquire about the buying habits of a certain sample population of buyers at brick-and-mortar stores. This is to determine whether people always look at their smartphones first before making a purchase.
This phase enables the researcher to figure out a conclusion out of the experiment. This must be based on rationality and logic in order to arrive at particular, unbiased outcomes. For example:
In the experiment, if a shopper consults first his or her smartphone before buying in-store, then it can be concluded that the shopper needs information to help him or her make informed buying decisions.
This phase involves the researcher going back to the empirical research steps to test the hypothesis. There is now the need to analyze and validate the data using appropriate statistical methods.
If the researcher confirms that in-store shoppers do consult their smartphones for product information before making a purchase, the researcher has found support for the hypothesis. However, it should be noted that this is just support of the hypothesis, not proof of a reality.
This phase is often neglected by many but is actually a crucial step to help keep expanding knowledge. During this stage, the researcher presents the gathered data, the supporting contention/s, and conclusions.
The researcher likewise puts forth the limitations of the study and his hypothesis. In addition, the researcher makes recommendations for further studies on the same topic with expanded variables.
VII. Advantages and Disadvantages of Empirical Research
Since the time of the ancient Greeks, empirical research had been providing the world with numerous benefits. The following are a few of them:
Empirical research is used to validate previous research findings and frameworks.
It assumes a critical role in enhancing internal validity.
The degree of control is high, which enables the researcher to manage numerous variables.
It allows a researcher to comprehend the progressive changes that can occur, and thus enables him to modify an approach when needed.
Being based on facts and experience makes a research project more authentic and competent.
Despite the many benefits it brings, empirical research is far from being perfect. The following are some of its drawbacks:
Being evidence-based, data collection is a common problem especially when the research involves different sources and multiple methods.
It can be time-consuming, especially for longitudinal research.
Requesting permission to perform certain methods can be difficult, especially when a study involves human subjects.
Conducting research in multiple locations can be very expensive.
The propensity of even seasoned researchers to incorrectly interpret the statistical significance. For instance, Amrhein et al. (2019) made an analysis of 791 articles from five journals and found that half incorrectly interpreted that non-significance indicates zero effect.
Source: Amrhein et al. (2019)
VIII. Samples of Empirical Research
Here are some notable examples of empirical research:
Since ancient times until today, empirical research remains one of the most useful tools in man’s collective endeavor to unlock life’s mysteries. Using meaningful experience and observable evidence, this type of research will continue helping validate myriad hypotheses, test theoretical models, and advance various fields of specialization.
With new forms of deadly diseases and other problems continuing to plague man’s existence, finding effective medical interventions and relevant solutions had never been more important. This is among the reasons why empirical research had assumed a more prominent role in today’s society.
This article was able to discuss the different empirical research methods, the steps for conducting empirical research, the empirical research cycle, and notable examples. All of these contribute to support the larger societal cause to help understand how the world really works and make it a better place.
Abbott, B., Abbott, R., Abbott, T., Abernathy, M., & Acernese, F. (2016). Observation of Gravitational Waves from a Binary Black Hole Merger. Physical Review Letters, 116 (6), 061102. https://doi.org/10.1103/PhysRevLett.116.061102
Akpinar, E. (2014). Consumer Information Sharing: Understanding Psychological Drivers of Social Transmission.
(Unpublished Ph.D. dissertation). Erasmus University Rotterdam, Rotterdam, The Netherlands. http://hdl.handle.net/1765/1
Altmetric (2020). The 2019 Altmetric top 100. Altmetric.
Amrhein, V., Trafimow, D., & Greenland, S. (2019). Inferential statistics as descriptive statistics: There is no replication crisis if we don’t expect replication. The American Statistician, 73, 262-270. https://doi.org/10.1080/00031305.2018.1543137
Dashti, H., Jones, S., Wood, A., Lane, J., & van Hees, V., et al. (2019). Genome-wide association study identifies genetic loci for self-reported habitual sleep duration supported by accelerometer-derived estimates. Nature Communications, 10 (1). https://doi.org/10.1038/s41467-019-08917-4
de Groot, A.D. (1969). Methodology: foundations of inference and research in the behavioral sciences. In Psychological Studies, 6. The Hague & Paris: Mouton & Co. Google Books
Doll, R., Peto, R., Boreham, J., & Isabelle Sutherland, I. (2004). Mortality in relation to smoking: 50 years’ observations on male British doctors. BMJ, 328 (7455), 1519-33. https://doi.org/10.1136/bmj.38142.554479.AE
Fairclough, N. (2003). Analyzing Discourse: Textual Analysis for Social Research. Abingdon-on-Thames: Routledge. Google Books
Gabriel, A., Manalo, M., Feliciano, R., Garcia, N., Dollete, U., & Paler J. (2018). A Candida parapsilosis inactivation-based UV-C process for calamansi (Citrus microcarpa) juice frink. LWT – Food Science and Technology, 90, 157-163. https://doi.org/10.1016/j.lwt.2017.12.020
Gallus, S., Bosetti, C., Negri, E., Talamini, R., Montella, M., et al. (2003). Does pizza protect against cancer? International Journal of Cancer, 107 (2), pp. 283-284. https://doi.org/10.1002/ijc.11382
Ganna, A., Verweij, K., Nivard, M., Maier, R., & Wedow, R. (2019). Large-scale GWAS reveals insights into the genetic architecture of same-sex sexual behavior. Science, 365 (6456). https://doi.org/10.1126/science.aat7693
Gonzalez-Morales, M. G., Kernan, M. C., Becker, T. E., & Eisenberger, R. (2018). Defeating abusive supervision: Training supervisors to support subordinates. Journal of Occupational Health Psychology, 23 (2), 151-162. https://dx.doi.org/10.1037/ocp0000061
Google (2020). The 2019 Google Scholar Metrics Ranking. Google Scholar
Greenberg, D., Warrier, V., Allison, C., & Baron-Cohen, S. (2018). Testing the Empathizing-Systemising theory of sex differences and the Extreme Male Brain theory of autism in half a million people. PNAS, 115 (48), 12152-12157. https://doi.org/10.1073/pnas.1811032115
Grullon, D. (2019). Disentangling time constant and time-dependent hidden state in time series with variational Bayesian inference. (Unpublished master’s thesis). Massachusetts Institute of Technology, Cambridge, MA. https://hdl.handle.net/1721.1/124572
He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep residual learning for image recognition. The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 770-778. https://doi.org/10.1109/CVPR.2016.90
Hviid, A., Hansen, J., Frisch, M., & Melbye, M. (2019). Measles, mumps, rubella vaccination, and autism: A nationwide cohort study. Annals of Internal Medicine, 170 (8), 513-520. https://doi.org/10.7326/M18-2101
Jamshidnejad, A. (2017). Efficient Predictive Model-Based and Fuzzy Control for Green Urban Mobility. (Unpublished Ph.D. dissertation). Delft University of Technology, Delft, Netherlands. DUT
Kamberelis, G., & Dimitriadis, G. (2011). Focus groups: Contingent articulations of pedagogy, politics, and inquiry. In N. Denzin & Y. Lincoln (Eds.), The SAGE Handbook of Qualitative Research (pp. 545-562). Thousand Oaks, CA: Sage. ISBN 978-1-4129-7417-2
Knowles-Smith, A. (2017). Refugees and theatre: an exploration of the basis of self-representation. (Unpublished undergraduate thesis). University College London, London, UK. UCL
Kulp, S.A., & Strauss, B.H. (2019). New elevation data triple estimates of global vulnerability to sea-level rise and coastal flooding. Nature Communications, 10 (4844), 1-12. https://doi.org/10.1038/s41467-019-12808-z
Levitt, H. M., Bamberg, M., Creswell, J. W., Frost, D. M., Josselson, R., & Suárez-Orozco, C. (2018). Journal article reporting standards for qualitative primary, qualitative meta-analytic, and mixed methods research in psychology: The APA Publications and Communications Board task force report. American Psychologist, 73 (1), 26-46. https://doi.org/10.1037/amp0000151
Long, J., Shelhamer, E., & Darrell, T. (2015). Fully convolutional networks for semantic segmentation. 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 3431-3440. https://doi.org/10.1109/CVPR.2015.7298965
Martindell, N. (2014). DCDN: Distributed content delivery for the modern web. (Unpublished undergraduate thesis). University of Washington, Seattle, WA. CSE-UW
Mora, T. (2019). Transforming Parking Garages Into Affordable Housing. (Unpublished undergraduate thesis). University of Arkansas-Fayetteville, Fayetteville, AK. UARK
Ng, M., Fleming, T., Robinson, M., Thomson, B., & Graetz, N. (2014). Global, regional, and national prevalence of overweight and obesity in children and adults during 1980–2013: a systematic analysis for the Global Burden of Disease Study 2013. The Lancet, 384 (9945), 766-781. https://doi.org/10.1016/S0140-6736(14)60460-8
Ogden, C., Carroll, M., Kit, B., & Flegal, K. (2014). Prevalence of Childhood and Adult Obesity in the United States, 2011-2012. JAMA, 311 (8), 806-14. https://doi.org/10.1001/jama.2014.732
Schenker, J., & Rumrill, P. (2004). Causal-comparative research designs. Journal of Vocational Rehabilitation, 21 (3), 117-121.
Shereen, M., Khan, S., Kazmi, A., Bashir, N., & Siddique, R. (2020). COVID-19 infection: Origin, transmission, and characteristics of human coronaviruses. Journal of Advanced Research, 24, 91-98. https://doi.org/10.1016/j.jare.2020.03.005
Sipola, C. (2017). Summarizing electricity usage with a neural network. (Unpublished master’s thesis). University of Edinburgh, Edinburgh, Scotland. Project-Archive
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2015). Going deeper with convolutions. 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 1-9. https://doi.org/10.1109/CVPR.2015.7298594
Taylor, S. (2017). Effacing and Obscuring Autonomy: the Effects of Structural Violence on the Transition to Adulthood of Street Involved Youth. (Unpublished Ph.D. dissertation). University of Ottawa, Ottawa, Canada. UOttawa