Practical Research Chapter 4

Research design

• Another term for research design is planning.

• Research design is what provides the structure for the procedures followed, the data collected, and the data analysis that is conducted.

Difference between research design and research methodology

• While there might be considered a “general approach” to planning a research paper, how a researcher goes about collecting and analyzing data is not considered the research design. This is the research methodology, which is often more specific towards different disciplines.

• This is why it is vital that when a researcher is planning their topic, they must also consider the kind of data their paper needs and how they will go about collecting it.

Primary data

• The most valid, truth-illuminating evidence that is found in a research paper. Derived directly from sources.

Secondary data

• Data derived from primary data. Interpretations made from the primary data.

Five guidelines for planning data collection

1. What data is needed?

   1. It is incredibly important to your research to know what kind of data you need to collect and its medium– is it documents, questionnaires, etc.?

2. Where is the data located?

   1. You must know precisely where you will be getting the data. Which library or research society will you be getting the data from? Plan ahead!

3. How will the data be obtained?

   1. Knowing where the data is isn’t enough; you need to know how to acquire the data with privacy and IRB laws, etc. Knowing how you will get your data separates your project into a viable one rather than just a dream.

4. What limits will be placed on the nature of acceptable data?

   1. Not all data is acceptable for use in a research project. Specific criteria and standards must be set up and met for the data to be used. Try not to be too vague or have partially completed data, etc.

5. How will the data be interpreted?

   1. MOST IMPORTANT QUESTION! How will you use the data to solve your research question or a sub-problem? Will the research lend itself to shed light on the problem— if not, you need to rethink your question

Quantitative research:

• Explains and, predicts, confirms TEST theories

• Focused approach

• Known variables

• Preplanned

• Established guideline

• Detached view (less bias)

• Numerical and standardized data from a large sample

• Stats analysis

• Very objective and nonbiased

• Communicated with numbers and statistics and a formal voice

Qualitative research:

• Describes and explains, explores, BUILD theory

• Holistic approach

• Unknown variables

• Flexible

• More personal

• Textual and image-based data

• Informative and small data collected

• Loosely structured and nonstandardized

• Searches for themes and categories

• Acknowledges bias

• Communicated with words, personal voice 

Mixed Methods

Answer different kinds of questions so you can better understand the world from various perspectives by combining them.

• You can count on quantifying certain kinds of data in a qualitative research approach

  • Makes me think of the cake paper last year; there was some form of numbers and scaling used in it

• Quantitative researchers can report on the feelings of the people they’re surveying

• Studies of human behavior frequently use a mixed-methods approach

Nine guidelines for deciding whether to use a quantitative or qualitative approach

1. Consider your comfort level with the assumptions of the qualitative approach. For example, if you think that no single reality is underlying your research problem and other individuals have different valid realities, then qualitative might be better than quantitative.

2. Ensure the audience you are creating the research paper for supports your approach; if the audience does not favor a qualitative study, it would be best to avoid that.

3. You must consider the nature of your question, as qualitative designs have advantages in addressing more interpretive research questions. Still, quantitative designs would be better for testing hypotheses and cause-and-effect relationships.

4. Reflect on how extensive the literature is because a topic with a weaker literature base will be more effective with a qualitative design, allowing more freedom in exploring phenomena and pinpointing variables affecting the topic.

5. The depth of what you plan to discover must be considered, as exploring phenomena with smaller groups is better for qualitative research. In comparison, larger groups are more effective with quantitative research.

6. The amount of time allotted is vital, as qualitative research takes a more extended amount of time than quantitative research. If you do not have much time, it might be better to stick to a quantitative approach.

7. Consider the extent to which you are willing to interact with the people in your study

   1. Qualitative researchers working with people need to built support and trust which takes more advanced planning and time

8. Consider the extent to which you feel comfortable working without much structure

   1. Qualitative researchers work with less guidelines and can be much more exploratory– be creative!

9. Consider your ability to organize and draw inferences from a large body of information

   1. Qualitative research requires a lot of field notes and large bodies of data to gather info from. Conducting a few statistical analyses is a lot easier than this

10. Consider your writing skills

    1. Qualitative researchers have to have really good writing skills to communicate all the data they have collected

Credibility

• Similar to worth. Is your research valid? Does it serve a purpose? I.e., is it credible?

Internal validity

• The extent to which its design and the data it yields allow the researcher to draw accurate conclusions about cause-and-effect and other relationships within the data.

Triangulation

• Multiple data sources are collected with the hope that they will all converge to support a particular hypothesis or theory.

Unobtrusive measure

• People are observed in a manner that they do not know they are being monitored. It must be noted that this could cause ethical issues due to observing individuals without their consent.

Thick description

• A researcher who uses thick description describes a situation in sufficiently rich, “thick” detail so that readers can draw their own conclusions from the data presented.

Respondent validation

• A researcher takes conclusions back to the participants in the study and asks quite simply, Do you agree with my findings? Do they make sense based on your own experiences?

Outliers/discrepant voices

• Researchers should actively search for examples that show inconsistencies with existing hypotheses. They should revise their theory until any examples are noted and accounted for.

External validity

• The extent to which its results apply to situations beyond the study itself

Representative sample

• Researchers attempting to learn more about a group of objects typically study samples from the group to create overall conclusions. For example, if a researcher is studying a particular snake in one region of the world, they would assume that what they concluded relates to that snake in other regions as well.

Assessment

• Collecting and analyzing data to understand the research question in a non-generalizable way that asks the people about stuff

  • Like a questionnaire or survey type of research or interviews even– similar to the LGBTQ+ Health paper where the students were asked or assessed on their views

Measurement

• Limiting the data of any phenomenon—substantial or insubstantial—so that those data may be interpreted and, ultimately, compared to a particular qualitative or quantitative standard.

Difference Between Substantial Phenomena and Insubstantial Phenomena 

• Substantial Phenomena

  • These are measurable things that have a physical presence in the real world. For example, an astronomer can measure the luminosity of light during a night sky. 

• Insubstantial Phenomena

  • These are measurable things that only exist as concepts or opinions; they are not physically measurable. One example can be seen when attempting to measure how much different students have learned. 

Scales

• Any type of measurement will fall into the categories of nominal, ordinal, interval, or ratio. These measurements are vital in determining what type of measurement will be used in processing data.

Nominal

• This is a method of organizing data into different categories through names. Assigning a category a name restricts your data to that name's definition. 

• An example of this would be categorizing a group of children into boys and girls, allowing the data to be analyzed based on the children’s gender.

Ordinal

• This method organizes data through terms such as greater than and less than. This allows data to be ranked based on one being higher than the other.

• An example of this would be levels of education, which are grouped into completion of elementary, middle, high school, bachelor's, and graduate school. These all have a clear order and ranking in which one can be considered “greater” than the other.

Interval

• This method of organizing data has two distinct features. The first is equal units of measurement, and the second is the zero point does not necessarily represent a complete absence of what is being measured.

• The temperature measurement Celcius (C) is an excellent example of this method, as there are equal intervals between the measurements, and there will not be any temperature at all if Celcius reaches 0.

Ratio

• This method of data organization is similar to the interval scale in that it has equal unit measurements. Where it differs is in the fact that there is an absolute zero point.

• A good example of the ratio scale is measuring length, as zero centimeters is the absence of any length, and there are equal unit measurements.

Validity

• Validity is the idea that the tool used actually measures what it is supposed to measure. This comes more into play with insubstantial phenomena, as there is nothing to base it on in the real world.

Face validity

• This is seeing if an instrument looks like it is measuring what it is supposed to measure. However, this form of validity is seen as subjective. This is seen more when confirming an individual's cooperation for a research study.

Content validity

• This validity measurement is about how well the form of measurement represents the content area. Content validity is often seen if any relevant topic of the domain is covered correctly to ensure a thorough assessment. 

Criterion validity

• This is the measure to see how well a test or assessment is scored in relation to other tests or examinations. For example, measuring a salesperson's effectiveness should relate to the sales and whether they increase.

Construct validity

• Construct validity is the measurement of characteristics that are unable to be physically observed. However, these behavioral constructs, like creativity and happiness, are assumed to exist due to patterns in behavior.

Reliability

• Reliability relates to how consistently an instrument can generate results when what is measured does not change.

Interrater reliability

• The measure of a judgment's consistency is when multiple individuals evaluate a specific topic.

Test-retest reliability

• This is the measurement of how well one instrument can gain the same results even when tested on multiple different occasions.

Equivalent -forms reliability

• This measures the ability of different variations of the same instrument to yield related results.

Internal consistency reliability

• This is the measure of how well all of the items in an instrument can yield a related result.

Standardization

• Standardization in research means that any instrument must be used with all individuals involved in the study. This ensures no changes in how an individual responds to your instrument because of errors in how it was presented to them. 

Ethical Issues in Research

• Protection from Harm

  • When conducting research, it is essential that the individuals involved do not come under physical or psychological harm. 

  • To prevent this, researchers must be sensitive to any topic that is potentially harmful to people, such as topics on self-harm or eating disorders.

  • Debriefing with participants is vital to ensure comfort and understanding of what they will be a part of.

• Voluntary and Informed Participation

  • Any research involving other humans requires informed consent, which means that those participating in the study must understand the nature of the research and be willing to grant written permission.

  • This can be done through an informed consent form, which briefly describes the study and what it plans to contribute. The form must also explain what participants will be asked to do, any possible risks, and the potential benefits the study will bring to not only the participants but the topic of research as well.

• Right to Privacy

  • Any individual who participates in a study has the right to privacy. Researchers should not state any person by name when discussing their findings.

  • In the modern day, it is also vital that researchers ensure that a participant's name or information can not be obtained through internet hacking. This can be ensured by not posting quickly decodable data. 

• Honesty with Professional Colleagues

  • When reporting on what they have done, researchers must show complete honesty in what they found. Nothing should be misleading about how said information was found, nor should any data be fabricated. 

  • It is also important to give credit to others when mentioning any information on their work and findings. Without proper acknowledgment, you are a thief, and your paper is unethical.

• Internal Review Boards

  • Any potential research must first be reviewed by the Internal Review Board (IRB). The IRB carefully inspects any proposal relating to human research. This board is made up of other researchers and scholars, and they ensure that the research will not cause harm to participants.

  • Any research that could potentially cause harm to animals is reviewed by the Institutional Animal Care and Use Committee (IACUC). 

  • While newer researchers view the IRB and the IACUC as hurdles to conducting their research, these boards are vital to ensuring ethical research is conducted.

• Professional Code of Ethics

  • Different disciplines will have their own code of ethics that must be followed when conducting research on humans or animals. 

  • Some examples of these organizations with their own code of ethics are the American Anthropological Association, the Society for Conservation Biology, and the American Psychological Association.