Research Methods Dersi 2. Ünite Özet

Introduction

Scientific research is the process of systematic and planned data collection, data analysis, assessment and interpretation, and reporting in order to figure out reliable and valid solutions to problems by limiting them. A scientific research process involves 1) problem identification, 2) data collection, 3) data analysis, 4) interpretation, and 5) report writing. All these stages are relevant parts of quantitative, qualitative, and mixed method research efforts. If one wants to make use of different methods, then s/he should consider each method’s unique specifications during planning stage.

Planning Quantitative Research Studies

Quantitative research is based on positivist paradigm, and it strives to formulate universal laws for each field of scientific study. Such a stance with respect to the nature of reality led to the birth of a research design to produce knowledge that is measurable, replicable, generalizable, intact, and value-free judgements. Unlike qualitative and mixed methods, quantitative studies entail a strict planning process that researchers have to abide step by step right from the beginning.

Identifying the Research Problem

The very first step of a research process is problem identification, which can be distilled from various sources including daily life, practical issues, findings of previous research, and relevant theories Besides, a researcher’s own experience and observations, discussions among scientists from the same field, and different opinions can also serve as sources to formulate research problems. After problem identification, research problem has to be defined. Research problem is often constructed as an interrogative statement that aims to analyze the relation between two or more variables. A well-defined quantitative research problem has to bear some certain features. First, the variables in the problem have to explain a relation. Second, it has to be formulated as a question. Third, a research problem has to be suitable for analysis based on collected data, or it has to be testable.

Primary building blocks of quantitative research are the variables. A variable can bear at least two different values. The term variable is used to refer to features that can change. Depending on their values and control types, variables are classified in two groups. With respect to their values, they are labeled as either discrete or continuous variables. As for the control types, variables are classified into three: dependent, independent, and control variables. Dependent variable is the one whose change is observed in comparison to other variables. Independent variable, on the other hand, is the one that has some kind of effect over the dependent variable. Control variables are those that can affect the dependent variable but not among the independent variables.

Once the problem is identified, the next step is to formulate a hypothesis, which indicates the researcher’s educated-guess about the result before any data is collected.

Selecting the Research Design

Research design is the arrangement necessary to collect and analyze research data in the most economical way. There are two primary research designs often utilized for quantitative research studies, which are survey and experimental designs. A researcher’s choice to employ either survey or experimental design is highly correlated with the hypothesis or research questions s/ he develops in the previous stage. If a researcher writes a hypothesis or research questions directed to assess an intervention’s or innovation’s effect over the result, s/he should stick with experimental designs. On the other hand, if a researcher aspires to describe participants’ attitudes, opinions, behaviors, and features as they are without any interference, then s/he should make use of survey models.

Survey

Survey serves to determine general tendency, attitude, or opinions of a population quantitatively or in numerical values by working on a sample chosen from that population.

Experimental Designs

Experimental designs provide the researcher with the opportunity to determine the independent variable’s effect, to assign subjects to different groups, and to assess behaviors and responses. It scientifically unravels what to do to either confirm or support an allegation.

The Classic Randomized Experimental Design

In this design, a researcher should foresee the formation of two groups: one (or more) is the experimental group consisting of participants given an experimental intervention, and the other is the control group composed of participants who are not either given any experimental manipulation or provided with any test stimuli.

Post-test Design

Post-test design also accommodates two variables (dependent and independent) and two groups. Again, subjects are randomly assigned to each group. Whereas the experimental group undergoes an intervention, the control group completes the research without any manipulation. Next, dependent variable is measured for each group. Lack of pre-test is the only difference between this design and classical randomized experimental design.

Repeated-Measurement Design

This design is preferred when the precise time for the emergence of independent variable’s effect is not certain and when it is not clear when to conduct the most reliable pre-test. Accordingly, researchers administer different assessment and measurement tools before and after the experiment.

Multiple-Group Design

In multiple-group design, more than one experimental and control group is formed to compare different levels of experimental variable. Use of this design is generally advised when the independent variable has more than one value or when a researcher wants to see the potential effects of manipulating the independent variable in different ways. Multiple-group designs may contain only one post-test as well as a post-test and a pre-test.

Randomized Field Experiments

Laboratory experiments cannot be used to examine some of the phenomena appealing to the interest of the political scientists despite their power to derive causal deductions. Yet, the foundational principles of experimental design can still provide guidance. A field experiment adopts randomization and variable manipulation by applying the techniques of experimental design to naturally-occurring situations and units. While randomly selected individuals or people clusters in this sense receive experimental intervention, the others form the control group. Following the completion of the experiment, the researcher conducts post-test measurements to estimate whether the intervention had an effect or not.

Quasi-Experiments

Quasi-experimental designs involve similar processes to those of classical experimental design. The only distinction regards how participants are assigned to groups.

Intervention Analysis

The dependent variable is subjected to pre-test and posttest procedures prior to testing the independent variable experimentally in this version of time series design, a.k.a intervention analysis or interrupted time series analysis. Just like the other non-randomized designs, formation of the independent variable is observed but not interfered by the researcher.

Identification of the Population and Sample

Population can be defined as the group to be associated with the results that will be distilled after data analysis. Research studies contain two types of populations: target population and accessible population. Target population is the one that is almost impossible to reach out. On the other hand, accessible population is the realistic choice of the researcher, and it is easily attainable.

Sample is a representative piece of the population whose features are examined. When the entire population is out of reach for a researcher, the proper size of a sample can be calculated via some statistical procedures or by examining the available sources. Finally, after deciding on the size (number of participants) of the sample, the researcher should consider which probability-based sampling strategy(ies) can be employed. Most often utilized sampling strategies are random sampling, stratified sampling, cluster sampling, systematic sampling, and convenient sampling.

Data Collection

All kinds of numerical values obtained via a method, technique, or instrument are accepted as data in quantitative research. Primary data collection techniques employed in quantitative research are tests and questionnaires. Tests are mainly utilized to measure personality, skills, success, and performance. Questionnaire, on the other hand, is a structured interview technique aiming to collect information about more than one characteristic of a person via various item types. Unlike tests, questionnaires are designed to obtain information about more than one feature of an individual.

Data Analysis and Interpretation

Analysis of quantitative data is processed through descriptive and inferential statistical techniques. Common descriptive analysis techniques often employed in quantitative research include frequency, percentage, arithmetic mean, weighted mean, mode, median, range, and standard deviation. Inferential statistical techniques, on the other hand, are of two types: non-parametric statistical techniques and parametric statistical techniques.

Researchers need a standardization to interpret quantitative research data, which is labelled as significance level. A significance level (or alpha level) is a probability level that reflects the maximum risk you are willing to take that any observed differences are due to chance.

Planning Qualitative Research Studies

Qualitative research strives to build up a theory by investigating and understanding social phenomena in their natural environments. During a qualitative research process, researchers focus on people’s behaviors, actions, and interactions. Research process aims to comprehensively understand human behaviors in their natural environments, which requires the researcher be flexible, re-arrange the research process in accordance with obtained data, adopt interpretive approach, and conduct data analysis through an inductive approach. A well-defined starting point or rigid procedures that a researcher should follow are not among the properties of a qualitative research process. During the planning stage, researchers should 1) form a theoretical framework as the foundational basis, 2) determine a systematic, manageable, and flexible research design, and, 3) turn the research into a meaningful, consistent, and reader-friendly report.

Identifying the Research Problem

Identifying the research problem is the very first step of a qualitative research process. Literature, researcher’s personal and professional experience, and social issues discussed in the field bear a crucial role in terms of identifying the research problem. Literature review includes reading and learning about theories, research studies, and documents, which eases problem identification. Personal and professional experiences can serve as fruitful background knowledge as they are distilled from a researcher’s own life and field of study. They can facilitate solving important problems via research. Lastly, the benefits that the research findings promise to the society make the problem meaningful.

Theoretical/Conceptual Framework

Theoretical framework eases defining different aspects of the research problem, developing data collection instruments, and forming themes to be employed during data analysis. Yet, it may not always be possible to postulate a theoretical background for all qualitative research endeavors. There is no theory that can explain an event or a phenomenon, a new theory can be developed based on the data collected through a qualitative research. One of the qualitative research designs, grounded theory is a main design serving to this end.

Developing Research Question(s)

In qualitative research, research questions are not formulated to include hypotheses and variables. The overall aim of qualitative research is to comprehensively understand perceptions and processes. Research questions generally begin with either how or what . A typical qualitative research question may be as follows: “ What are individuals’ perceptions about refugees?”

Selecting Research Design

Qualitative research designs guide researchers in terms of determining research focus, data collection instruments, and data analysis techniques. Qualitative research designs provide researchers with a flexible approach and maintain consistency across different stages of the research process in line with a specific focal point. Sometimes, the research plan can be reviewed and even re-structured depending on emerging conditions during the process. Frequently employed research designs in qualitative approach are action research, phenomenology, case study, ethnography, narrative inquiry, and grounded theory.

Selecting the Participants

Purposeful sampling strategies are employed for participant selection in qualitative research. Quantitative endeavors mostly employ probability-based sampling strategies. Primary purposeful sampling strategies used in qualitative research are extreme or deviant case sampling, maximum variation sampling, homogeneous sampling, typical case sampling, critical incident sampling, snowball or chain sampling, criterion sampling, confirming and disconfirming case sampling, and convenient sampling.

Data Collection

Observation, interview, and documents are main data collection techniques in qualitative research approach.

Data Analysis, Reporting Findings, and Interpretation

There are several approaches that qualitative researchers can follow during data analysis. Some of these approaches include thematic analysis, narrative analysis, discourse analysis, semiotic analysis, descriptive analysis, and content analysis. The researcher tries to unfold the meanings and patterns embedded in the data set through her/his perspective. There are some principles that can guide researchers in analyzing research data, reporting findings, and interpreting. These principles are 1) coding the data, 2) forming the themes, 3) arranging and identifying the data in accordance with codes and themes, and 4) reporting the findings.

Limiting the Results and Making Analytical Generalizations

Qualitative research does not aim to make generalizations because it works with limited number of participants in a comprehensive manner and focal social events have a dynamic and changeable nature.

Planning Mixed Method Research Studies

Mixed method is a research approach where both quantitative and qualitative data sets are gathered and integrated so as to find answers to research questions. In this approach, gathered data sets are analyzed separately and findings are interpreted collectively after collecting both types of data simultaneously.

The Title

The title of a well-developed mixed method research should include the research subject and sample. Besides, the phrase mixed method should also be used in the title to signal the employed research method.

Identifying the Research Problem

While explaining the research problem, one should avoid merely listing relevant studies in the literature. Even though citing studies conducted on a similar or same research problem is inevitable and crucial, researchers should also write about what should be done. They should come up with more reasons

Identifying Research Aim and Questions

In general, the sentences denoting the aim of a mixed method research are comprehensive and long. Mixed method research questions should be developed after stating the aim. A mixed method research includes (i) hypothesis or quantitative research questions (ii) qualitative research questions, and (iii) mixed method research questions. Mixed method research questions could be developed via either method-centered or contentcentered approaches. But, the best mixed method research questions are those that reflect both the method and   content. Such research questions are also known as hybrid mixed method research questions.

Selecting Mixed Method Research Design

Different designs are employed to explain how the research process is built in a mixed research method. These designs include The Convergent Design, The Explanatory Sequential Design, The Exploratory Sequential Design, The Intervention Design, The Social Justice Design, and The Multistage Evaluation Design

Sample

Because quantitative and qualitative approaches are simultaneously employed in a mixed method research, the plan should be to use probability and purposeful sampling techniques for the quantitative and qualitative aspects respectively. The best one that could be utilized for the quantitative aspect is random sampling. Besides, simple random sampling, stratified sampling, or multistage cluster sampling can also be appropriate depending on the research aim

Data Collection

Due to the fact that mixed method research entails using both quantitative and qualitative approaches, data collection should be planned in two stages. Accordingly, qualitative data collection tools such as interview, observation, and document analysis and quantitative tools such as questionnaire, test, and scales can easily be employed for mixed method research. Yet, the mixed method design determined in accordance with the research questions dictates which data collection technique will be employed and in which order.

Data Analysis, Interpretation, and Reporting the Results

Because mixed method design makes use of both quantitative and qualitative research methods simultaneously, previous explanations as to how to conduct data analysis applies here as well.

Diagrams Used in Mixed Method Research

Several diagrams are employed to clarify the procedures and underline the significant points within mixed method research. These diagrams display data collection, data analysis, and interpretation of results.