The Importance of a Focused Research Strategy

Introduction

Frequently, employees of a company seek to know everything there is to know about their goods, facilities, systems, and so on. The research strategy is determined by the knowledge we need to gather in order to make significant decisions about a good, service, program, or other subject (Oxley, Rivkin and Ryall, 2010). Typically, then you’ll be confronted with a big decision because of existing consumer concerns, the need to persuade funders / bankers to lend funds, unmet customer demands, or the need to improve an internal mechanism (ROYNE, 2008). The more concentrated you are on what you hope to learn from your study, the more accurate and profitable the research can be, the less the time it can take, and the least it can cost you. In terms of the scope and complexity of knowledge you get, there are trade-offs as well. Typically, the more range you like, the less scope you’ll find. In the other side, whether you try to look at a certain feature of a commodity, operation, software, or something in great depth, you won’t likely receive too much knowledge about other facets (ROYNE, 2008). For those just getting started in research or with minimal capital, a variety of approaches may be used to obtain a reasonable balance of scope and depth of knowledge. They will learn more about specific aspects of their goods, systems, schemes, and so on without going bankrupt in the process.

Main Research Paradigms in the Field Of Management and Research

It is a system within which all scientific thought and ideas are organized. The best examples of “a pre-paradigmatic study field,” are entrepreneurship and management (Freshwater and Cahill, 2012). It goes without saying that a well-chosen management analysis paradigm can be an effective method for performing research (Machado and Davim, 2020). Management methodology paradigms are study methods that have varying views on the test’s reality, the knowledge that can be used in the report, the role of principles, and the data collection techniques that should be included (Kuzmin, 2019). Furthermore, administrators tend to follow common objectives and produce different outcomes while utilizing different paradigms when doing analysis.

The positivist model, realism paradigm, interpretivist paradigm, and pragmatism paradigm are the four primary management science paradigms (Saunders, Lewis, and Thornhill, 2009). The positivist paradigm is characterized as a study approach that focuses on finding knowledge from a singular point of view, without making references to other topics (Somekh and Lewin, 2005). To put it another way, only primary material that can be observed is included, with no citations to other authors (Morgan, 2007). The information realism is gathered primarily by quantitative methods and is highly organized with a large number of tests (Burkhardt and Virely, 2015).

The realism paradigm is characterized as a study approach that aims to find the reality by any means necessary while maintaining a sense of modesty toward the conclusions reached (Owen, 2006). The evidence gathered from other researchers, as well as societal and world perspectives, are both taken into consideration by the writer (Vaughan, 2012). The interpretivist methodology is a quantitative analysis approach utilized in social science that focuses on the human person and the techniques he or she employs to interpret facts (Immy, 1997). The primary goal of this type of study is to examine and discover something different without relying on a large number of instances (Freshwater and Cahill, 2012). This type of study is often linked to previous studies, but the conclusions are often new.

The Difference between Quantitative and Qualitative Research

Quantitative analysis focuses on figures and information while gathering and interpreting data, while qualitative study focuses on terms and their definitions (Bryman, 2006). Both are essential for the acquisition of various types of information.

Quantitative Research

Numbers and graphs are used to represent quantitative analysis. It’s used to put hypotheses and conclusions to the test or to validate them. This research approach is beneficial for developing generalizable facts about a subject (Jervis and Drake, 2014). Experiments, percentage-based assumptions, and surveys of closed-ended questions are all traditional quantitative methods. The aim of a quantitative study review is to create a relation within a population between one thing an independent variable and another a contingent or outcome variable (Carl Chiarella., Reiner Franke., Peter Flaschel. and Willi Semmler., 2006). Quantitative research designs may be descriptive (subjects are usually only evaluated once) or experimental (subjects are assessed both before and during a treatment).

A descriptive survey may only define a relationship between variables; causality may only be established in an experimental analysis. Statistics, logic, and a neutral point of view are all aspects of quantitative research. Over divergent reasoning, quantitative analysis stresses numerical and reliable proof, as well as detailed, convergent reasoning. The usage of jargon is a distinguishing feature of qualitative research. It’s a tool for understanding thoughts, theories, and emotions. This method of learning helps you to have a deeper understanding of topics that aren’t well-understood. Qualitative methods include open-ended interviews, written explanations of findings, and literature reviews that explore theories and theories.

Quantitative and Qualitative Research’s Strengths and Weaknesses

Qualitative Analysis

Quantitative research have information that can be interpreted numerically, thus the term. We may use mathematical tests to make conclusions regarding the data when the data is in numerical form. Statistical statistics such as mean, median, and standard deviation, as well as inferential statistics such as t-tests, ANOVAs, and regression study correlations, are examples of these statistics (MRC). Statistical analysis enables useful knowledge, such as interest trends, population differences, and statistics, to be extracted from study findings (Camayd-Freixas and Donahue, 1987). Multivariate statistics such as the MRC and stepwise association analysis break down the data much more to decide what variables, such as variations in interests, may be attributed to differences within particular populations, such as age groups. Quantitative experiments mostly utilize automated data collection techniques, such as polls, but we may often use static methods, such as two-alternative, forced-choice studies or comparative benchmarking to examine error rates and time on mission (Nielsen, 2011).

The great strength of quantitative research is that they have descriptive data—for example, helping us to take a glimpse of a consumer population—but we have trouble interpreting them (Ismail, 2020). Gallup surveys, for example, often include statistics on the President of the United States’ popularity ratings, but they don’t provide the critical detail we’d like to analyze the data. Approval poll data in quantitative form Gallup’s president popularity survey contains quantitative results. It’s impossible to determine whether voters support or disapprove of President Obama’s job performance in the absence of the evidence that will be used to analyze these presidential job approval figures. Some respondents may believe President Obama is too liberal in his conduct, whereas others may believe he is too traditional, but there is no way to know without the requisite evidence.

Although the sample size may be used to manipulate a p-value, a large enough sample size is needed to provide enough statistical ability to decide if a result is correct. Even if your findings are right, if your analysis is underpowered due to a limited sample size, you may not be able to obtain statistical significance. If you gain statistical significance for a limited sample size, on the other side, you don’t need to maximize the sample size; the result is valid regardless. While a small sample scale allows determining anything more complex, once you can decide anything with a small sample size, the result is just as valid as if you used a broad sample size.

This data deficit in a product-development setting may result in crucial errors in product design. For example, a survey may reveal that the majority of users like 3D displays, leading a product team to decide to include one in their product. However, unless the majority of users choose autostereoscopic 3D displays—that is, 3D displays that do not need the wearer to wear glasses—or 3D displays that are only used for watching sports or action movies on television, using a 3D monitor with glasses for data processing on a handheld device might not be the right design direction. Furthermore, such a study should only be carried out by someone who is well-versed in the use and interpretation of quantitative statistics. Many experiments place so much emphasis on the p-value and sample size. The p-value is a figure that shows how likely it is that the results of a study are due to chance. The observations are said to be statistically important if the p-value is less than.05, indicating that there is less than a 5% probability that the effects are due to chance (Bryman, 2006).

Qualitative Research

The attributes or characteristics of things are defined using data from qualitative studies. You won’t be able to quickly convert these definitions to figures, like you would for quantitative analysis results, but you can do so using an encoding process. Qualitative analysis experiments will provide you insights into human nature, emotion, and personality traits that quantitative study cannot. Qualitative data provides knowledge regarding consumer preferences, needs, interests, habits, usage cases, and a host of other factors that are critical in creating a product that fits into a recipient’s life.

Rather than doing a mathematical study after collecting data, analysts check for patterns in the data. When it comes to determining patterns, analysts aim for statements that are the same across all participants in the study. Hearing a comment from only one participant is anecdote; hearing it from two is a coincidence; and hearing it from three is a pattern, according to the rule of thumb. Brand growth, management decisions, and marketing campaigns will also be influenced by the patterns you discover.

Since you can’t use mathematical tests to test these patterns, you can’t use a p-value or an impact size to validate them as you can for quantitative statistics, so use them with caution. Furthermore, such data can be checked on a regular basis by a qualitative analysis program. With enough time and money, you will partake in a process known as behavioral coding, which entails assigning numeric identifiers to qualitative behaviour and converting it into quantitative data that can then be analyzed statistically. Behavioral coding allows you to do a series of additional analyses, including latency temporal analysis, which is a mathematical test that distinguishes sequences of behavior, such as those for Web site navigation or job workflows. Using behavioral coding to code the findings, on the other hand, takes a long time and costs a lot of money. Furthermore, only the most highly educated researchers are normally able to code conduct. As a result, this strategy is usually prohibitively expensive.

Furthermore, since qualitative data collection cannot be automated as easily as quantitative method, gathering vast volumes of data, as is characteristic of quantitative analysis studies, is typically highly time intensive and costly. As a result, qualitative study often involves just 6 to 12 participants, while quantitative research often involves hundreds or even thousands of participants. As a consequence, when it comes to finding and verifying patterns, qualitative analysis continues to have fewer observational capacity than quantitative research.

Incorporating Quantitative and Qualitative Research

While both these and analysis methods have advantages and disadvantages, they can be highly successful when used in tandem. You may use qualitative analyses to identify the factors that influence the areas under consideration, and then use that information to design quantitative research that assesses how these factors could influence customer preferences (Bryman, 2006). To return to our earlier example of display preferences, if qualitative analysis had identified the display type—for example, tv, computer screen, or cell phone display—researchers may have used that information to generate quantitative research that would enable them to determine how these variables could affect customer preferences. Simultaneously, you will incorporate patterns discovered by quantitative analysis into qualitative data-collection processes, allowing you to confirm the trends.

While this may seem to be the polar opposite to what we’ve just said, it’s actually a really simple process. Younger users favor autostereoscopic screens only on handheld devices, whereas older users prefer conventional displays on all devices, as an indication of a qualitative pattern. You may have figured this out by posing an open-ended, qualitative query like, “How do you feel about 3D displays?” This topic may have prompted a debate regarding 3D displays, with the distinctions between stereoscopic, autostereoscopic, and conventional displays being clarified. In a corresponding comparative survey, you might ask a set of questions like, “Rate your degree of preference for a standard 3D view on a mobile device—which includes 3D glasses—with options varying from strongly prefer to strongly dislike,” with options ranging from strongly prefer to strongly dislike. An automated machine assigns a numerical value to each alternative selected by a participant, enabling a scientist to easily collect and interpret vast quantities of data.

Different Techniques of Data-Gathering Operationalised

The method of collecting and assessing information on variables of interest is a systematically specified way of answering requests, specifying study questions, testing theories and evaluating results. The total amount of data produced and processed worldwide is expected to exceed 149 zettabytes by 2024 (Hallinan, 2019). There are, however, several explanations for collecting data, but we will concentrate here mainly on advertisers and small business owners relevant. Data is the starting point for any investigation. Measurement, counting, and observation are some of the methods used by investigators to gather evidence. The methods and procedures used to gather data are primarily determined by the study’s objectives. We must always ask ourselves the following questions if we decide to gather data:

• What data do we need to gather in order to address the study questions raised by our research objectives? This applies to the variable’s selection.
• What method would we use to gather this information? This applies to the research style that was chosen.
• What are the locations where we want to gather information? The community that would be sampled is referred to as this.
• What data collection methods and technologies can we employ? This is a term used to describe a method of gathering data.
• What percentage of the populace would we be looking into?
• How many topics will we cover in our investigation? This applies to the number of people in the survey.
• What method would we use to choose our sample? This is a term that applies to the sample style.

The Data Collection Techniques

Data may be collected in a multitude of forms, ranging from a simple observation at one location to a large-scale survey spanning a whole country in another part of the world. The method of data collection selected would have a significant effect on how well the data is collected. Structured surveys, questionnaires, observational forms, checklists, and tape archives are only a few of the devices used to collect evidence. Data collection strategies allow us to obtain information about our study subjects (individuals, objects, and occurrences) as well as the sense in which they exist in a comprehensive manner. We must gather data in a methodical manner. If data are compiled haphazardly, conclusive responses to our research questions would be difficult to come up with. The same measuring unit, operational definition of variables, metric, and so on must be used in the entire data collection. Data may be collected in a variety of ways. The method chosen is primarily determined by:

1. Time, resources, and staffing are all factors to consider.
2. Study design
3. Study objectives

Qualitative Data Methods

Life History

The usage of life history is a particular use of the interview method. This method enables people to share stories that reveal what they believe to be significant about their viewpoints. The life history method is a unique interviewing methodology that uses a small sampling size of no more than 25 participants. This method is well-suited to conservative rural communities. The pattern of fertility, as well as women’s thoughts regarding marriage, childbirth, and abortion, are all issues that lend themselves well to research utilizing the life-history method.

Case Studies

A case study entails a thorough examination of a small group of individuals, a family, a neighborhood, or a specific circumstance. It is a qualitative research project. A case study, according to Young, is a way of looking at and assessing the existence of a social entity, whether it’s an individual, a family, an organization, a cultural group, or even an entire society. Several techniques of data collection are usually employed at the same time. Non-probability selection is often used to select research subjects. For instance, the cases may be chosen to be representative or illustrative of a specific phenomenon or category.

Service Statistic

Both national family planning agencies compile reports on their services. A management information system has been developed in some organisations (MIS). However, the standard of service figures vary significantly across countries and also within countries. As a result, care should be exercised while using them. Service figures are often used by researchers to determine the conditions of the issue they wish to investigate. They may be used to equate the findings of a single report to national statistics in certain situations. Frequently, in operations analysis programs, supplemental types are used to include evidence that is not accessible from the standard service statistics.

Data Collection Methodology Using Key Informants

The key-informant Interview (KII) method of needs evaluation is based on the assumption that some people know a lot about the environment and the target demographic. It’s often assumed that these people will correctly express neighborhood desires in order to help in program preparation. In certain cases, key informants are interrogated in a semi-structured fashion (Hassan and Singh, 2012). The individual or individuals chosen as key informants must have a thorough understanding of the city, its social system, the facilities that would be given, and its residents. It’s a great way to resurrect facts regarding historical incidents or lifestyles that are no longer visible. The needs assessment’s goals will aid in determining the best kind of individual to use as a main informant. Elected representatives, long-time citizens, corporate owners, administrators, religious leaders, and people representing a range of lifestyles, genders, ideologies, or ethnic backgrounds may be considered by the researcher. Since few individuals in a group will comment on anything, the issue can be prioritized until the informant is chosen. Working with key informants may result in a number of approaches. Questions should be planned ahead of time, and they must be well organized.

Number of Approaches to Assess the Accuracy of Interview Results.

A researcher can start asking or more questions that provide the same sort of knowledge on purpose. The first question may be posed at the start of the interview, while the second may be asked at the end (Hassan and Singh, 2012). The responses to the two questions are then compared to see if they are consistent. This is one method of determining the data’s dependability. The interviewers will be told to probe for tough topics, important questions, or questions that the researchers want to make sure the knowledge is right. That is, the research can pose the same query in a slightly different way or echo the respondent’s response before asking whether the knowledge is right.

Field managers can be used to assist interviewers in stressful circumstances and to ensure that they are carrying out their responsibilities. In certain trials, one supervisor is assigned to every five interviewers.

Any experiments that use an interrogation protocol tend to reinter view a certain proportion of the participants. Re-interviewing around 5% and 10% of the participant, depending on the size of the sample, is a common practice (Rudolph, 2000). The first interview’s data is then compared to the second interview’s data to ensure that they are consistent. If there are significant discrepancies, particularly on issues such as age, marital status, and parity, there is obviously a problem. It’s possible that the issue is with the questionnaire, the interviewers, the tabulation processes, or something else entirely. It is necessary to do statistical tests for inconsistencies or accuracy after the data has been compiled and tabulated. A frequency distribution of women’s parity, for example, may show that a number of women appear to have 18 or 19 living children. Since this is very impossible, the investigator must decide whether to delete these surveys, exclude the details from the parity question, or re-interview the people who claim to have 18 or 19 kids.

Issues of Validity and Reliability

The consistency of answers to several coders of data sets is referred to as reliability in qualitative analysis. It can be improved by taking extensive field observations, tracking equipment, and transcribing digital data. Validity in qualitative analysis, on the other hand, can be described differently than it is in quantitative research (Gibbon, 1998). Internal validity, external validation, reliability, and objectivity are naturalists’ equivalents of internal validation, external validation, durability, and objectivity. In qualitative analysis, credibility, honesty, transferability, dependability, and confirmability are all factors that contribute to trustworthiness. Long involvement in the field and the triangulation of data points, procedures, and investigators to develop legitimacy are needed to operationalize these concepts. A detailed explanation is needed to ensure that the findings can be transferred between the study and the people being examined. Rather than searching for authenticity in qualitative analysis, studies strive for dependability, recognizing that the conclusions would be prone to adjustment and uncertainty. When it comes to qualitative analysis, Instead of using the term validity, it is established criteria such as systemic corroboration, consensual validation, and referential adequacy as evidence for claiming qualitative research’s legitimacy. The scientist utilizes a variety of data points to affirm or refute the explanation of systemic corroboration. It is the identification of four forms of validation triangulation, build validation, face validation, and catalytic validation.

Validity and Reliability in Quantitative Research Paradigm

Experimental tests and objective measurements are used in quantitative research to evaluate theories, and generalizations are the results of this test (Lee, 2000). They often place a strong emphasis on determining and analyzing causal correlations between variables (O’Dwyer and Bernauer, 2008). Scientists described the quantitative paradigm of study as follows (SÜRÜCÜ and MASLAKÇI, 2020): Charts and graphs depict the research’s findings, and commentators use terms like “variables,” “populations,” and “effect” in their everyday vocabulary and though we don’t really understand what each word means (Wachtel, 2015). However, we are aware that this is an inevitable aspect of the testing phase. The term “analysis” has become synonymous with “quantitative studies” in the public domain.

Researchers using the qualitative paradigm seek to categorize phenomena into observable or universal definitions that can be generalized to all topics (Lee, 2000). Quantitative researchers’ primary prerequisite is the design of instrument(s) and their administration in a systematic manner centered on predetermined procedures (Patton, 2015). However, the issue is whether the measurement device is capable of measuring what it claims to be capable of measuring (O’Dwyer and Bernauer, 2008). The authenticity of an instrument is on the spotlight in the broadest way. The most crucial aspect of the study is to guarantee its reliability and validity (Roberts, 2015). “The degree to which findings remain stable over time and a correct reflection of the overall population under analysis is considered to be reliable, and if the results of a study may be replicated under identical methods, then the test instrument is assumed to be reliable,” according to the definition. The stability of the instrument can be calculated by utilizing divided half, test-retest, or parallel type approaches to ensure that the questionnaire items score stays consistent. When dealing with a constant measure, you’ll get the same answers every time. The degree of stability is proportional to the degree of reliability; a higher degree of stability leads to a higher degree of reliability, which indicates that the outcomes are repeatable.

The researcher may improve the research instrument’s repeatability and therefore reliability by enhancing its internal logic. To increase reliability, the researcher may update or exclude questionnaire [test] elements during this phase; however, this may jeopardize the instrument’s validity, and the key issue here is how much the revision changed the table of specifications. The validity of an analysis determines when it correctly estimates what it was supposed to measure, as well as the precision of the test results. The truth of quantitative analysis is referred to as “construct validity” (Elvik, 2008). The construct refers to the query definition, idea, or theory that the researcher uses to gather data and devise sampling plans that are compatible with the construct.

Validity and Reliability Assurance

As results are fully utilized, they become more reliable and defensible, leading to generalizability, which is one of the principles is the structure for both conducting and reporting high-quality qualitative study (Bashir and Marudhar, 2018). As a consequence, the accuracy of a study is linked to the generalizability of the findings and, as a result, to the checking and improvement of the study’s validity or trustworthiness (Mustafa, 2021). Validity relates to the degree of congruence between the theories of phenomena and the facts of the universe. There is disagreement about the naming of specific concepts; some terms that may be seen in this context include reflexivity and expansion of results. To respond to the issue of how to improve validity, they said that refining sampling and data collection strategies in the data collection phase improves validity (BULUT KILIÇ, 2016). To be generalizable is a component that specifically differentiates quantitative and qualitative analysis approaches.

In qualitative studies, the study has used multi-method methods to increase the generalizability of the analysis, which improves the research’s reliability and validity (Reynolds, 1988). When a researcher invests adequate time in the field and uses various data collection methods to corroborate his or her conclusions, researcher prejudice may be reduced (BULUT KILIÇ, 2016). Many researchers agreed that convergence is a popular method for enhancing the reliability and validity of analysis or findings assessment.

Sampling

Sampling is a technique for selecting people or a subset of the population in order to make statistical inferences and estimate the target population. Different screening methods are often used in market studies such that researchers may not have to study the whole population in order to obtain actionable perspectives. It’s also a time- and money-saving technique, so it’s used as the basis for all study designs. In a research survey program, sampling methods can be used for optimum derivation. For example, if a drug company wishes to explore a drug’s adverse side effects on the whole population of a country, conducting a clinical study that involves anyone is almost impossible (Robertson, 2017). In this case, the researcher chooses a sample of people from each demographic and studies them, supplying him or her with preliminary information on the drug’s effectiveness. There are several different kinds of sampling techniques, which can be divided into two groups: likelihood sampling and non-probability sampling. As a consequence, all qualified individuals would have a higher chance of being chosen for the survey, and you will be able to more accurately generalize the study’ results. Probability sampling procedures take longer and are more expensive than non-probability sampling techniques. Since non-probability (non-random) selection doesn’t really start with a full sample size, certain people have just a small likelihood of being chosen.

Procedures for Testing

Systematic sampling

Random sampling is more complicated and time-consuming than systematic sampling. It may also make it easier to cover a large study range. Systematic sampling, on the other hand, adds some subjective parameters into the results. This can lead to an over-representation and under in such patterns (Madow, 1953). Because of its convenience, systematic sampling is widespread among researchers. If a random trait appears excessively for any “nth” data set, researchers typically believe the findings are indicative of most typical populations which is unlikely (Patton, 2015). To start, a researcher chooses an integer as the starting point for the scheme. This percentage must be less than the population as a whole for example, for a 100-yard football field, they don’t sample every 500th yard. Following the selection of a number, the researcher selects the distance, or spaces between samples in the population.

Simple Random Sampling

The estimation of sampling error and the elimination of selection bias are also possible for stratified random sampling, as for other chance sampling methods (Yadav and Pandey, 2013). It is the simplest form of chance sampling, which is a distinct benefit (Greenwood, 2007). One disadvantage to non-probability sampling is that you may not be able to find enough people who share the desired trait, particularly if it is unusual. It may also be difficult to establish a complete sampling frame and inconvenient to contact them, particularly if you need several channels of interaction (email, cell, and telegraph) and the sample units are distributed over a wide geographic area.

Stratified Sampling

When conducting study or examination on a community of individuals with similar characteristics, a researcher may discover that the population size is too big to conduct research on (Michalcová, Lvončík, Chytrý and Hájek, 2011). An observer can take a more practical approach by choosing a small group from the population to save time and resources (ZHOU and YE, 2013). A sample size is a small proportion of the population that is used to reflect the whole population (Bryman, 2019 p. 123). The stratified random sampling approach is one of several methods for selecting a sample from a population. The whole community is divided into homogeneous classes called strata in a stratified random sampling (plural for stratum). After that, random samples are chosen from each stratum (Saini and Kumar, 2018). Take the case of an independent researcher who wants to know how many MBA students got work offers within three months of graduation in 2007.

Clustered Sampling

Cluster sampling, also known as one-stage cluster sampling, is a technique for detecting and incorporating into a sample clusters of participants that make up the community (Sutradhar, 2020). The method of finding a grouping of people who make up the population and including them in the sample party is known as cluster sampling (Olea, 2007). This is a popular method for conducting market research (Thompson, 2013). The primary objectives of cluster sampling are cost savings and improved sampling accuracy. This technique may also be used in combination with multi-stage sampling (Seber and Salehi, 2013). The fact that in cluster sampling, a cluster is considered as a sampling unit, while in stratified sampling, only unique elements of strata are accepted as sampling units, is a significant distinction between cluster and stratified sampling (Tokola, Larocque, Nevalainen and Oja, 2011). As a result, the sampling frame of cluster sampling is made up of a full list of clusters. The source of primary data is then selected at random from a few clusters.

Conclusion

Many higher degree research students and even early career researchers find it difficult to formulate and incorporate the principle of study paradigm in their research proposals. The current paper uses an ethnographic and hermeneutic approach to clarify the significance of study paradigm, drawing on our several years of practice as Research Methods lecturers as well as relevant literature. The paper clarifies the core facets of study paradigms that researchers should be familiar with in order to properly address this idea in their research proposals. It includes advice about how researchers should place their study inside a model and how to justify their choice of paradigm.it is therefore important to have an accurate data that will make the output analysis reliable.

Reference List

Bashir, J. and Marudhar, M., 2018. Reliability & Validity of the Research. Scientific Journal of India, 3(1), pp.66-69.

Bryman Alan, Emma Bell, and Bill Harley.Business Research Methods. Fifth edition. Oxford, United Kingdom ; New York, NY: Oxford University Press, 2019.

Bryman, A., 2006. Integrating quantitative and qualitative research: how is it done?. Qualitative Research, 6(1), pp.97-113.

BULUT KILIÇ, İ., 2016. Validity and Reliability Study for Studio Work Course Time Management Scale. Journal of Educational Sciences Research, 6(2), pp.113-129.

Burkhardt, K. and Virely, S., 2015. Real Business Cycle Theory and Critical Realism (Transcendental Realism). Journal of Critical Realism, 14(3), pp.287-305.

Camayd-Freixas, Y. and Donahue, M., 1987. Quantitative Analysis of Curriculum Effectiveness. [Place of publication not identified]: Distributed by ERIC Clearinghouse.

Carl Chiarella., Reiner Franke., Peter Flaschel. and Willi Semmler., 2006. Quantitative and empirical analysis of nonlinear dynamic macromodels. Bingley, U.K.: Emerald.

Elvik, R., 2008. The predictive validity of empirical Bayes estimates of road safety. Accident Analysis & Prevention, 40(6), pp.1964-1969.

Freshwater, D. and Cahill, J., 2012. Paradigms Lost and Paradigms Regained. Journal of Mixed Methods Research, 7(1), pp.3-5.

Greenwood, J., 2007. Review: Strange Bedfellows Fiona J. Hibberd, Unfolding Social Constructionism. New York: Springer, 2005. 207 pp. ISBN 0387229752 (hbk). Theory & Psychology, 17(4), pp.605-607.

Hallinan, D., 2019. Opinions ∙ Data Protection without Data: Could Data Protection Law Apply without Personal Data Being Processed?. European Data Protection Law Review, 5(3), pp.293-299.

Hassan, A. and Singh, R., 2012. Respondents versus Informants Method of Data Collection: Implications for Business Research. SSRN Electronic Journal,

Hibberd, F., 2001. Gergen’s Social Constructionism, Logical Positivism and the Continuity of Error. Theory & Psychology, 11(3), pp.323-346.

Jervis, M. and Drake, M., 2014. The Use of Qualitative Research Methods in Quantitative Science: A Review. Journal of Sensory Studies, 29(4), pp.234-247.

Kuzmin, S., 2019. Methodology of paradigms in organizational growth studies. Upravlenets, 10(5), pp.52-62.

Lee, R., 2000. Book Review: Mixed Methodology: Combining Qualitative and Quantitative Approaches. Field Methods, 12(3), pp.256-258.

Machado, C. and Davim, J., 2020. Research methodology in management and industrial engineering. Cham: Springer.

Madow, W., 1953. On the Theory of Systematic Sampling, III. Comparison of Centered and Random Start Systematic Sampling. The Annals of Mathematical Statistics, 24(1), pp.101-106.

Michalcová, D., Lvončík, S., Chytrý, M. and Hájek, O., 2011. Bias in vegetation databases? A comparison of stratified-random and preferential sampling. Journal of Vegetation Science, 22(2), pp.281-291.

Morgan, D., 2007. Paradigms Lost and Pragmatism Regained. Journal of Mixed Methods Research, 1(1), pp.48-76.

Mustafa, C., 2021. Qualitative Method Used in Researching the Judiciary: Quality Assurance Steps to Enhance the Validity and Reliability of the Findings. The Qualitative Report,.

Nielsen, R., 2011. Cues to Quality in Quantitative Research Papers. Family and Consumer Sciences Research Journal, 40(1), pp.85-89.

O’Dwyer, L. and Bernauer, J., 2008. Quantitative research for the qualitative researcher.

Olea, R., 2007. Declustering of Clustered Preferential Sampling for Histogram and Semivariogram Inference. Mathematical Geology, 39(5), pp.453-467.

Oxley, J., Rivkin, J. and Ryall, M., 2010. The Strategy Research Initiative: Recognizing and encouraging high-quality research in strategy. Strategic Organization, 8(4), pp.377-386.

Patton, M., 2015. Qualitative research & evaluation methods. Thousand Oaks, Calif.: SAGE Publications, Inc.

Reynolds, D., 1988. British school improvement research: the contribution of qualitative studies. International Journal of Qualitative Studies in Education, 1(2), pp.143-154.

Roberts, A., 2015. A Principled Complementarity of Method: In Defence of Methodological Eclecticism and the Qualitative-Quantitative Debate. The Qualitative Report,.

Robertson, D., 2017. Side effects and adverse drug reactions. Nurse Prescribing, 15(10), pp.512-514.

ROYNE, M., 2008. Cautions and Concerns in Experimental Research on the Consumer Interest. Journal of Consumer Affairs, 42(3), pp.478-483.

Saini, M. and Kumar, A., 2018. Ratio estimators using stratified random sampling and stratified ranked set sampling. Life Cycle Reliability and Safety Engineering, 8(1), pp.85-89.

Seber, G. and Salehi, M., 2013. Adaptive Sampling Designs. Berlin, Heidelberg: Springer Berlin Heidelberg.

Sutradhar, B., 2020. Multinomial Logistic Mixed Models for Clustered Categorical Data in a Complex Survey Sampling Setup. Sankhya A,.

SÜRÜCÜ, L. and MASLAKÇI, A., 2020. Validity and reliability in quantitative research. Business & Management Studies: An International Journal, 8(3), pp.2694-2726.

Thompson, S., 2013. Sampling. Hoboken, N.J.: Wiley.

Tokola, K., Larocque, D., Nevalainen, J. and Oja, H., 2011. Power, sample size and sampling costs for clustered data. Statistics & Probability Letters, 81(7), pp.852-860.

Vaughan, B., 2012. Pierpaolo Donati,Relational Sociology: A New Paradigm for the Social Sciences. Journal of Critical Realism, 11(2), pp.255-261.

Wachtel, M., 2015. Charts, Graphs, and Meaning: Kiril Taranovsky and the Study of Russian Versification. Slavic and East European Journal, 59(2), pp.178-193.

Yadav, S. and Pandey, H., 2013. A Ratio-cum-Dual to Ratio Estimator of Population Variance Using Qualitative Auxiliary Information under Simple Random Sampling. Mathematical Journal of Interdisciplinary Sciences, 1(2), pp.91-96.

ZHOU, H. and YE, J., 2013. Stratified sampling particle filter algorithm based on clustering method. Journal of Computer Applications, 33(1), pp.69-71.