Research Methods: Data Types and Sets

• The presentation covers research methods and techniques.

Data Types

• Cross Sectional Data is data collected by observing many subjects at a single point in time.
• Panel Data, also known as repeat cross sectional data set, studies how subjects change over a period of time, observing the same subjects at different times.
• Time Series Data is collected over a period of time, often taken at successive, equally spaced points, and is used for trend analysis and forecasting.

Primary and Secondary Data Sets

• Primary Data is generated by the researcher through surveys, interviews, and experiments designed to solve a particular research problem.
• Secondary Data uses existing data from government institutions and healthcare facilities as part of organizational record keeping.
• HIES data set is one example of secondary data.

Comparison of Primary and Secondary Data Sets

• Primary data is first-hand data gathered by the researcher, while secondary data is data collected by someone else earlier.
• Primary data is real-time data, whereas secondary data is past data.
• Primary data collection is more involved than secondary data collection.
• Primary data is collected through surveys, experiments, questionnaires, and interviews, while secondary data comes from government publications, websites, books, and journal articles.
• Primary data collection is expensive and time-consuming, but always specific to the researcher's needs, available in crude form, and more accurate and reliable.
• Secondary data collection is quick, easy, and economical, but may or may not be specific to the researcher's needs, available in refined form, and relatively less accurate.

Determinants of Sample Selection in Scientific Study

• Entire populations can't be studied, so a sample must be selected.
• Samples need to be representative of the population.
• Sample Size depends on the time and resources available.
• Larger populations need larger samples.
• More heterogenous populations need bigger samples.
• Acceptable Sample Sizes are generally between 100-1000, but this is not a hard rule.

Concepts and Variables

• Concepts are the building blocks of theories, used to grasp real-world phenomena, and they define logical relations.
• Concepts are subjective and hard to use in research study due to their nature.

What are Concepts

• Kumar (2000) noted concepts are mental images with varied value from person to person.
• Concepts are subjective impressions, and their understanding will differ from person to person, which, if measured, could cause problems in comparing responses.
• Concepts need to be converted to variables so that they can be measured on different scales with different precision.
• If a researcher uses concepts, indicators of these concepts need to be identified subjectively but logically linked.
• These indicators can then be converted into variables.
• Adjustment to university, including academic, social, and psychological factors, can be one such general concept.

Difference Between Concepts and Variables

• Measurability is the main difference.
• Variables are measurable with varying degrees of accuracy.
• A variable consists of two components: a statistical unit and a property.
• A variable's value varies from individual to individual.
• Variables can be numerical or qualitative.
• Examples for adjustments in university are academic grades and submitting assignments on time.
• Social variables include sports participation and friendships.

Types of Variables

• Variables can be independent, dependent, extraneous, or intervening.
• From a causation view point, an independent variable is one that affects the dependent variable and isn't affected by the dependent variable.
• One example is student grades being a dependent variable that relies on independent variables such as age, gender, study hours, attendance, and class participation.

Dependent Variable

• A dependent variable relies upon a set of independent variables.
• Good grades depend on age, gender, study hours, attendance, class participation, and being a hostel/day scholar.

Extraneous Variable

• Reality isn't perfect with the variables, there are extraneous ones.
• Extraneous Variables impact the dependent variables aside from the independent variables.
• In labs, its easier to control extraneous variables.
• In a study of memory, age is an extraneous variable.

Intervening Variables

• In certain situations an intervening variable is needed to affect the dependent.
• This variable is not constant, but plays an important role causing cause and effect.

Example

• It is generally considered that a rich person has a higher life expectancy than a poorer person. The intervening variable is what causes this?

Likert Scale

• The Likert Scale is a scale used to represent people's attributes.

Operationalization

• Operationalization means turning abstract conceptual ideas into measurable observations.
• Social anxiety isn't observable but can be defined in terms of self-rating scores relating to crowded avoidance and anxiety symptoms. It is measuring a phenomenon.
• A researcher may wish to measure the concept of "anger," which is intangible.
• Other measures, such as facial expression, vocabulary, loudness, and tone of voice may be used as indicators of "anger" by outside observers.
• The most direct way to measure the depth of anger is to ask "are you angry?" or "how angry are you?", but this is problematic because everyone understands the concept differently.
• Loudness might be uniform, but some might respond verbally while others respond physically.
• An example could be job satisfaction as the independant and job turnover as the dependent variable.

Researchers Achievements

• Research Accounts can be found on Research Gate and Google Scholar.

Research Gate

• On Research Gate, the researcher uploads all its publications after creating an account, but also downloads.
• One can send a message for restricted papers.
• Access that week's stats for publications from around the globe.
• Has weekly citations, total research score, coauthors scores, etc.
• Can check, follow publications, researchers and scientists.

Google Scholar

• Creating a profile on Google Scholar allows one to view total citations, individual citations, h index, i10 index, and coauthor details.
• Year-wise tracking of research paper citations is possible.

Scopus Author Profile

• The Scopus author profile is considered more reliable
• It only consists of papers and citations.
• Total publications can be viewed.
• Info around citations can be viewed.

LinkedIn

• Individual profiles and organizations appear on LinkedIn.
• LinkedIn profiles get viewed.
• One can add achievement paragraphs, publication details, a background and experience.
• Employment is an option.
• Relationships can be built with people.