Data Analysis Seminar 5

Data Analysis Seminar 5

• Seminar presented by Dr. Waqas Ahmed
• Focuses on quantitative and qualitative data analysis
• Location: Nizhny Novgorod

Getting Data Ready for Analysis

• Data coding involves assigning numbers to participant responses for database entry
• Data entry is the process of inputting coded responses into a database, using software like SPSS
• Raw data can be entered via other software

Editing Data

• Outlier responses are significantly different from other responses
• Inconsistent responses don't align with other information
• Illegal codes are values not in the coding instructions

Transforming Data

• The presentation shows a screenshot of SPSS software
• Data are organized in rows and columns.
• Data manipulation and analysis (such as computing or using formulas) are shown within the software. Various variables and data points from the database are visible.

Getting a Feel for the Data

• Explains scale types for data analysis
• Presents various methods for visually summarizing data, including bar charts, pie charts, histograms, scatterplots, and others. Tables (Table 14.1) visually categorize scale types by data examples, analysis methods and displaying the summary

Frequencies

• Analysis of data within SPSS software
• Presentation of frequency tables
• Data visualization through graphical representations
• Computation of descriptive statistics (mean, median, mode, etc.)
• Shows functions and tools available for data manipulation with examples

Descriptive Statistics - Central Tendencies and Dispersions

• Presents SPSS software; data arranged in tables; shows data manipulation tools for central tendency and summary statistics

Reliability Analysis

• Shows SPSS software, with a focus on reliability analysis tools
• Includes tools for various statistical tests, model exploration and analysis.
• Examples of variables and statistical functions

Quantitative Data Analysis: Hypothesis Testing

• Focuses on hypothesis testing in quantitative data analysis
• Discusses the one-sample t-test, a technique to evaluate whether the mean of a population matches a standard
  • Shows one-tailed and two-tailed tests
  • Explains the rejection and acceptance regions for different hypothesis tests
• Discusses "Analysis of Variance" (ANOVA) as a technique to analyze differences in means across more than two groups of an interval- or ratio-scaled dependent variable

Regression Analysis

• Simple regression analysis analyzes the relationship where one independent metric variable affects a dependent metric variable

Statistical Significance Tests

• Illustrates different statistical tests (Fisher's Exact Test, Students T-Test, Regression Analysis, ANOVA)
• Graphs accompany each test explaining the visual representation of the tests and the data being analyzed (X and Y-axis)
• The example data are related to evaluating characteristics of a tumor and drug dose.

Conceptual Model

• Presentation of a model illustrating relationships between variables with arrows to depict interaction, directions and influence.
• Includes ideas of moderating effects (like gender, race, etc), mediating variables, independent and dependent variables

Hypothesis based on conceptual model

• Outlines hypothesized relationships and effects (Main Effect, Moderation, Mediation, Multiple predictor and Interaction)

Model Validation

• Explains the components of a model validation, like face validity, statistical validity (model fit, parameter/model significance, effect strength, discussion of multicollinearity from correlation matrix) and predictive validity (out-of-sample issues).

Step 1

• Model Summary - shows adjusted R squared values, and estimates for step one
• ANOVA shows the model and residual data and figures
• Coefficients data with B, Std.Error, Beta, t and P values

Step 2

• Model Summary - illustrates adjusted R squared values, and estimate for step two
• ANOVA table - data about the model and residual data.
• Coefficients data with B, Std.Error, Beta, t, and p-values

Step 3

• Model Summary - shows adjusted R squared values, and estimates for step three
• ANOVA shows the model and residual data and figures
• Coefficients data with B, Std. Error, Beta, t and p-values
• Highlights that some independant variables have an insignificant effect on the dependent variable while the other have a significant effect

Software for Quantitative Data Analysis

• Lists software options (SPSS, Minitab, Stata) for conducting quantitative analysis
• Provides links for each software program.

Qualitative Data Analysis

• Defines qualitative data as data that can't be easily measured numerically; that focus on attitudes, opinions, behaviours and experiences
• Describes the way qualitative data is collected
• Describes processes and procedures to explain, interpret and understand people and situations

Qualitative Data Collection

• Identifies methods for collecting qualitative data
• Includes: Observations, Interviews, Documents, Focus Groups, Audio/Video Recordings and Questionnaires

Coding

• Coding is a data reduction method to group similar data into manageable units.
• Helps visualize relationships and patterns in qualitative data.

Categories/Themes

• Explains the process of categorizing and creating themes from qualitative data, based on similar participant responses
• Shows that themes can be named based on data or by researcher observations
• Discusses different types of themes (ordinary, unexpected, hard-to-classify, major, and minor)

Organizing Data

• Discusses methods for organizing qualitative data sets
• Includes: Reviewing all data, grouping key data into smaller groups, material organization by type (observation, interview, field notes)

Exploring Data

• Explaining data to get a overall sense of the data set.
• Reading through the data
• Ensuring data is complete and understandable before proceeding to analysis
• Summarizes each data point to give an initial sense of the data
• Taking notes and creating memos.

Steps in Coding the Data

• Steps involved
• Defining and assigning codes
• Making lists of code words and grouping them for relationships
• Returning to the data
• Summarizing codes into major themes

Identifying Themes

• Explains how to identify patterns and repeated events/phrases in data
• Focuses on themes and keywords within the data

Themes

• Explains different types of themes (ordinary, unexpected, hard-to-classify, major, and minor)

Summarizing Your Data

• Explains how to summarize qualitative data and its themes
• Importance of summing up learnings. Synthesis across sources

Major and Minor Themes from Teacher's Interview

• Shows how interviews with teachers can be analyzed.
• Shows how questions and responses from teachers can be grouped into major and minor themes
• Example questions and themes/subthemes

Collating Data into a Table of Coded Responses

• Shows how collected data and responses can be organized for analysis;
• Illustrates connections based on the major and minor themes and interviews collected

Explanation of Themes

• Discusses how to explain data and themes using quotes from participants

Example of Narrative Format

• Illustrates how to create a narrative explanation based on participants' quotes

Software for Qualitative Analysis

• Provides software options for qualitative data analysis (ATLAS.ti, NVIVO)
• Gives links for the softwear mentioned