Research Methods & Statistics

Phases of Research Process

• The empirical phase involves data collection and preparation for analysis.
• This phase consumes the most time.
• This phase involves gaining results, sorting them, and evaluating them.
• The result of this phase can be qualitative or quantitative, and either analogue or digital.

Levels of Measurement

• Nominal: Data is named, like eye color.
• Ordinal: Data is named with a natural order, like satisfaction levels.
• Interval: Data is named with a natural order and equal intervals, such as temperature.
• Ratio: Data is named with a natural order, equal intervals, and a true zero, like height.

Criteria for Assessing Quality of Measurement

• Reliability.
• Validity.
• Sensitivity.
• Objectivity.

Reliability

• Defines the ability of an instrument to create reproducible results.
• Similar scores should be obtained each time the tool is used.
• A questionnaire yields similar answers repeatedly.

Validity

• Defines whether an istrument measures what it is supposed to measure.
• Addresses whether the questionnaire provides answers to the intended research questions.
• Determines if the appropriate tool is being used.

Sensitivity

• Defined as the probability of correctly identifying a condition or disease state.
• Sensitivity is a statistic used to describe the accuracy of an instrument making a dichotomous classification.
• Calculated based on the relationship of the outcome of the test/instrument and the actual state of affairs.

Objectivity

• Ensures that everyone follows the same rules without subjective interpretation.

Assessment of Qualitative Data

• Trustworthiness.
• Credibility.
• Dependability.
• Confirmability.
• Transferability.

Trustworthiness

• Aims at addressing the subjective nature of qualitative research.
• Responds to the concerns of outsiders, by ensuring confidence in the study's outcomes.
• Ascertains reader's belief in reported findings.

Credibility

• Refers to confidence in research findings' truth.
• Assessed via purposive sampling ensuring participants possess similar knowledge and experiences on the study's phenomenon.

Dependability

• Shows the consistency and repeatability of the findings.
• Requires constant revisions with assistance from advisors, critics, and participants.

Confirmability

• Indicates the degree of neutrality.
• The extent to which study findings are shaped by respondents, not the researcher's biases or motivations.

Transferability

• The likelihood study findings hold meanings in similar situations.
• Requires member checks to confirm shared experiences and meanings among participants.

Statistics

• Definition: A branch of mathematics that summarizes, organizes, analyzes, and interprets numerical data.
• Improves data quality through experiment design and survey sampling and provides tools for prediction and forecasting.
• Applied across various academic disciplines, including natural and social sciences, government, business, and nursing.

Kinds of Statistics

Descriptive Statistics

• Statistical methods summarize or describe a collection of data.
• Organizes and summarizes numerical data from the population and sample.
• Measured.
  • Condenses data in frequency distribution with scores tested high to low.
  • Graphic presentation makes frequency distribution data apparent.
• Measures of central tendency describes the mean, median, and mode.
• Descriptive statistics analyze data to help describe and summarize data so patterns emerge and data can presented meaningfully for simpler interpretation.

Inferential Statistics

• Concerned with population and the use of sample data to predict future occurrences.

Inferential Statistics - Uses

• Estimating population parameter.
  • Sampling error is the the difference between data obtained from a random sample and measuring an entire population; sample does not accurately reflect population; may occur.
  • Sampling distribution is a theoretical frequency distribution based on an infinite number of samples which researcher never draws.
  • Sampling bias: occurs when samples aren't carefully selected, as in non-probability sampling.
• Testing the null hypothesis.

More Inferential Statistics

• These are techniques that allow the samples to make generations about the population from which the samples were drawn.
• Arises from sample incurring errors and is not expected to represent the population.
• Use decision theory.

Decision Theory

• Based on assumption associated with the theoretical normal curve, it tests differences between groups with expectations that all groups are members of the same poplulation.
• Expressed as a null hypothesis and the level of significance (alpha) is set at 0.05 before data collection.
• Cites there are 2 options when researcher is diciding what the result of statistical test means (Burns & Groove, 2007).

Type 1 Error

• Occurs when the null hypothesis is rejected when in reality is not.
• This happens when the level of significance is at 0.05 than at 0.01.
• Decreased when the level of significance become more extreme.

Type 2 Error

• Occurs when null hypothesis is regarded as true, but actually false.
• Statistical analysis indicates no significant differences between groups when the reality is they are different.
• There is a greater risk of error when the level of significance is 0.01 than when it is 0.05.

Power Analysis

• Used to control type 2 error as it determines the probability of the statisical test to detect significance.
• The research determines the sample size, the level of significance and the effect size on the outcome variable (Cohen, 1988).

Degree of Freedom

• Interpretation of a statistical test depends on numbers of values that can vary.
• Focus lies on the values that are not free to vary.
• Expression is generally Df sign and a number that denotes significant level (eg. Df =0.01 of 0.05).

Frequency Distribution Types

• Grouped Frequency Distribution: Used with nominal data or when continuous variables are examined, such as age, weight, blood pressure, etc.
• Ungrouped Frequency Distribution: This happens when data are categorized and presented in tabular form to display all numercial values obtained for a particular variable.
• Percentage Distribution: Percentage of subjects in a sample whose scores fall into a speific group and the number scores in that group.

Statistical Tools for Treatment of Data

• Percentage: Computed to determine proportion of part to whole.
• Ranking: Order of decreasing or increasing magnitude of variables.
• Weighted Mean: The overall average or responses/percertions of study responsents.
• T Test: Compares repsonse of two respondents groups in the study.
• ANOVA: Tests the differences between 2 mean which can be used to examind data from two or more groups.
• Factor Analysis: Examines relationships amound large numbers of varibles and isoalte those relationships to identify clusters.
• Regression Analysis: Used too predict the value of one variable when the value of variables are known.

Research Output

• Multiple Regression Analysis: Used to correlate more than two variables and the complete randomized block design.
• The design is the same as ANOVA except that complete blocks are used instead of items.
• Writer should know not only the parts in research process but also the forms and style in writing.

Preliminary Pages

• Title Page: A phrase describing the research study that should not not be too long/short or too vague/general.
• Endorsement Page: States the study has be examined and recommended for oral exam.
• Approval Sheet: Presents study has been approved by the Committee on oral examination.
• Acknowledgement Page: Researcher's deep gratitude for people who assisted and helped.
• Dedication: Allows the researcher to to personally dedicate.
• Table of Contents: Contains all the parts of the resource.

Other types of pages

• List of Tables: Follows the table of content and indicates the title of the tables.
• List of figures: Composed of paradigms, diagrams, graphs and charts or flowcharts.
• Abstract: Short summary of the completed research.

Abstract Details

• Should be self-contained and concise and explaining the research study as briefly and clearly as possible.
• Consists of concise statements of:
  • What the study is all about.
  • The methodology.
  • The most important findings.

Main Body

Chapter 1

• Introduction: Refers to what the study is about or what makes the researcher interested in the study; purpose is too introduce the reader to the subject matter.

Chapter 2

• Methodology.

Chapter 3

• Results and Discussions

Chapter 5

• Conclusions and Recommendations
• Answer sub problems, summary of study and conluding remarks that highlight thoughts, revision of the plan or improvement, if necessary.

Satisfy the following questions

• Did the intervention work?
• What should be changed?
• What should be the next step?

Supplementary Pages

• Bibliography.
• Appendices.

Data Analysis, Interpretation, and Presentation

• Involves qualitative analysis, quantitative simple and qualitative analysis, tools, theoretical, presenting findings.

Why Analyze Data

• Analyze data to obtain usable and useful information irrespective of the data is qualtitiative or quantiatative like describe, identify, compare.

Scales of Measurement

• Confusion exists about analysis type and forms of pectoral presentation or data display, but decision are based on scales.

Scales are as follows

• Nominal, ordinal, and numerical.

Nominal Scale

• Data can be classified into non-numerical categories, and the order in which these categories can be written or asked is arbitrary.

Ordinal Scale

• Data can be classfied into non-numerical which an inherent exists among the response categories.
• Ordinal scales observed call ratings of qualtiy/agreement.

Numerical Scale

• Numbers represent the possible response categories where there is a natural ranking of categories.

Common Myths

• Complex analysis and big words impress people.
• Analysis comes at the end.
• Quantitative analysis is the most accurate type.

Organizing the Data

• Organize all forms/questionaires.
• Check the accuracy.
• Assign a unique identifier to each form/questinaire.

Enter Data

• By hand/computer.

Computer Data

• Excell; spreadsheets.
• Microsoft Access.
• Quantitative analysis.
• Counts.
• Percents.
• Mode.
• Range.
• Standard.
• Ranking.
• Crosstab.

Interpreting Information

• Judgements needs to be made.
• Involve key stakeholders.
• Interpret what you learned like what was new or expected.
• Know is further study is needed.

Discuss Limitations

• Written explicitations and oral reports need to be prepared.
• Do not cause causation without a true experiment.
• Do not generalize population without random sample and quality administration like survey.

Qualitative Analysis

• Process of bringing order, structure and meaning to data which is messy and ambiguous.

_Marshall and Rossman, 1990:111 “Hitchcock and Hughes take this one step further: " ...the ways in which the researcher moves from a description of what is the case to an explanation of why what is the case is the case."

Simple Qulaitative

• Unstructured.
• Structred.
• Semistrucutued.
• Patterns.
• Data categorization scheme may be used.
• Critical incidents.

Tools to Support Data Analysis

Spreadsheet, statisical packages, tools with analysis tools.

Theoeretical Frameworks for Qualitative Analysis

• Basing data analysis around theoretical frameworks.
• Grounded theory, Distributed Cognition, Theory of Activity.

Grounded Theory

• Aims to derive theory from systmatic analysis that occurs through 'coding'.
• Identify category, flesh out links to subcategories and form theeoretical scheme.

Distributed Cognition

• People, environment and artifacts are regarded as one cognitive system used for analyzing collaborative work.
• Focuses on information propagation and information.

Activity Theory

• Explains human behavior in forms of pricatial activity in the world.
• Provides a framework that focuses analysis around the concern of 'activity' helps.

Models

• Activity
• Action
• Opertion

Summary

• Quantitative and qualitative data can be done on the data gathering that was done and various approaches.
• Percentanges and averages can be done using Interaction Design.
• Theory of Activity is to support data analysis.

This quiz covers key concepts in research methodology and statistics, including dependability, confirmability, transferability, measures of central tendency, statistical prediction, and sampling error. Also covers the use of descriptive and inferential statistics in research.