
About Course
The course very gradually, from simple to complex, immerses professionals from non-technical sciences (management, business, humanities, linguists, psychologists, sociologists, cultural scientists, economists, political strategists, etc.) in the exciting analysis of data and the search for hidden patterns and methods of predictive analytics – and will help you easily navigate, use and not be afraid.
The course is also suitable for engineering professionals who have not studied data analysis, but want to understand it – without unclear formulas and cumbersome calculations.
The course is based on the most modern materials demonstrating the possibilities of using the SPSS program in various fields (marketing and sociological research, personnel research, opinion polls, development of psychodiagnostic tools and tests, analysis and forecasting, etc.)
What Will You Learn?
- Loading and importing data from different sources
- Transforming and cleaning data, preparing the array for analysis
- Descriptive statistics: mean, mode, median, quartiles, etc.
- Predictive analytics
- Finding differences between groups
- Identifying hidden relationships between variables
- Classification tasks (will not give credit, buy will not buy goods, etc.), construction of neural networks
- Time series analysis, search for patterns and forecasting
- Syntax Basics
- Other SPSS Features
Course Content
Introduction
-
Author’s word
-
Just came to inquire for fun?
-
Warning before the course
-
How will we learn
Not about spss: basic non-technical concepts
-
Introduction to the section
-
Descriptive and analytical statistics
-
The Importance of Models in Analytics
-
Without a model: exploratory data analysis (rad) and data mining
-
Sample and population
-
Sample size calculation
-
Complex case: representativeness and sample size
-
Data array
-
Analysis objects (strings)
-
Variable characteristics
-
Types of scales for variables
-
Data type for variable values
-
Requirements for writing values in an array
-
Array Awareness
-
The concept of a statistical hypothesis
-
Formulation of hypotheses
-
Probability of error and significance level
-
Normal distribution
-
Параметрика и непараметрика
-
Functional and probabilistic relationships
-
Data analysis process in an organization
-
Section results
-
Testing on the results of the section
Introduction to spss
-
Preparing the workspace: installing tools
-
Basic interface elements
-
Bookmarkview data
-
TabView Variables
-
Invisible element: context menu
-
The ribbon is the main control part of the program (+ quick access panel)
-
Analysis results window: output
-
Your own language: syntax window
-
Section results
-
Interface
-
Get to know the interface yourself
Basics for a quick start
-
Introduction to the section
-
Entering variables: creating an array passport
-
Passportization
-
Data entry: variable values
-
Massiveness
-
Reverse: loading the finished array and editing variables
-
The essence of preparing data for analysis
-
Readiness
-
The essence of data analysis
-
Analyzer
-
The essence of data visualization
-
Visualizer: not just tables
-
Exporting analysis results
-
Export
-
Section results
Preparing data for analysis: loading, cleaning and transformation
-
Introduction
-
Extracting and Loading Data: Reading and Importing
-
Loading a finished array
-
Editing Variables
-
Adjust the variables yourself
-
Ordering Variables
-
Let’s move it around
-
Removing variables (columns)
-
Delete unnecessary
-
Displaying Variable Values
-
Two sides of the same coin
-
Filtering Selecting observations (rows) for analysis
-
(un)natural selection
-
Repairing samples using observation selection
-
Array splitting: automatic selection of observations in the context of virtual groups
-
Split
-
Sorting observations (rows)
-
Sorter
-
Removing rows
-
Lowercase “annihilation”
-
Detection and cleaning of duplicates
-
Duplication
-
Detecting and dealing with input errors
-
Primitive: a simple way to find errors
-
Detecting input errors in a meaningful way
-
Smyslovukha
-
About working with missing values
-
Missing values: casescasesobjects (strings)
-
Gaps in observations of objects (rows)
-
Missing values: variables (columns)
-
Passes on variables (columns)
-
Descriptive and analytical statistics
-
Ranking
-
Multiple Answers: Virtual Variable
-
Virtual set
-
Transpose an array
-
Transposition (switching columns and rows)
-
Weighing data: working with an aggregated array
-
Weigher for the unit
Descriptive Statistics
-
The essence of descriptive statistics
-
Frequency analysis (frequency distribution)
-
Interesting frequencies
-
Frequencies for several variables: contingency tables (cross-classification)
-
Interesting frequencies: crosstabs
-
Frequencies for Multiple Choice Responses
-
Many to unity
-
Turf analysis
-
Turfology
-
4 groups of measures in descriptive statistics
-
Measures of central tendency: mean, mode, median
-
Pp-qq
-
Comprehensive data explore
-
Review approach
-
Metrics ratio
-
Percentages
-
Custom Pivot Tables reports (custom tables)
-
Summary
-
Visualizations: diagrams
-
Visual
-
Section results
Analytical statistics: what is it?
-
Difference from descriptive statistics
-
3+1 main blocks of analytical tasks
-
Return to basics: model, rad and data mining
-
Back to basics: hypotheses, margin of error and significance
-
Checking the distribution of variables for compliance with normal distribution
-
Check for normal distribution: outdated windows for any version of spss
-
Section results
Analytical statistics: differences between groups
-
What is it used for?
-
Parametrics and non-parametrics again
-
Once again about the significance of differences between groups
-
Dependent (paired, related) and independent samples
-
Basic group comparison tool: contingency tables and chi-square test
-
Group Comparison: Chi-Square
-
Comparison of 2 independent groups (samples): t-test, parametrics
-
T-test for means (parametrics)
-
Comparison of 2 independent groups (samples): nonparametrics
-
Non-parametric for comparing 2 independent samples: mann-whitney
-
Lots of independent
-
Comparison of Groups in Analysis of Missing Values
-
Comparison of 2 paired (related) groups/samples: t-test, parametrics
-
Pair parametrics
-
Comparison of 2 paired (related) groups/samples: nonparametrics
-
Paired nonparametrics
-
Comparison of multiple paired samples
-
Paired set
-
Section results
A special case of group comparison: one-sample comparison tests
-
One-Sample Comparison Tests
-
T-test for means (parametrics)
-
Parameter for one sample
-
General overview of the window with one-sample tests for nonparametrics
-
Binomial test (non-parametric)
-
Binomial
-
Chi-square test (non-parametric)
-
Square chi
-
Testing the shape of the distribution (nonparametrics)
-
Ks-norm
-
Wilcoxon signed (median) test
-
Wilcoxon sign
-
Is the sequence of values random or not (non-parametric)
-
Exchange
-
Section results
Analytical statistics: relationships between variables
-
Causality and dependent and independent variables
-
Reasoning about relationships between variables
-
And once again about what a significant statistical relationship is
-
The essence of correlation of variables
-
Contingency tables again: just to check the connection
-
Communication via crossstabs
-
Checking the contact form
-
Form
-
Correlation analysis: strength, direction, significance
-
The strength of the connection is not important
-
Dealing with spurious correlations
-
Recognize a lie
-
The essence of regression analysis
-
Factor analysis
-
Factorization
-
Reliability-consistency (suitability) analysis
-
Suitability
-
Multidimensional scaling
-
Mnogomerka
-
Results of the section on searching for hidden relationships between variables
Analytical statistics: object classification
-
What is it used for?
-
Logistic regression
-
Logistics
-
Multinomial logistic regression
-
Multiple Logistics
-
Ordinal regression
-
Order
-
Probit analysis
-
Try
-
Discriminant
-
Decision trees classifications
-
Tree
-
Roc classifier (receiver operating characteristic)
-
Roc
-
Neural Networks: Multilayer Perceptron (mlp)
-
Neuronka
-
Results of classification from a “bird’s eye view””
Analytical Statistics: Basics of Time Series Forecasting
-
Introduction, purpose and subject area disclaimers
-
The main pitfall of forecasting over time
-
Main tasks of time series analysis
-
Time series components: trend, seasonality, cycle, burst
-
The starting point for trend analysis is to look at the chart by eye”
-
Start of time
-
Time series analysis: trend line with forecast and forecast “corridor”
-
Trendovik
-
““manifestation” of a trend using a moving average
-
Slip
-
Seasonal decomposition
-
Season
-
Auto- and cross-correlations
-
Temporal correlation
-
Section results
Syntax Basics: Introducing the Internal Spss Language
-
What is syntax and what can be useful for a business user
-
Opening, populating, launching and saving the syntax window
-
Rename variables and delete variables
-
Removing variables and changing their names
-
Calculation of variables (compute)
-
Calculation
-
Conditional statements if, and and or
-
Conventions
-
Recoding variables (recode with to, into and else)
-
Translation
-
Basic aggregator functions (sum, mean, count, min, max)
-
Aggregations
-
Comments in syntax (* or /*)
-
Commentator
-
Why take the syntax directly from the user interface and be able to edit it?
-
Section results
A brief overview of the individual capabilities of spss and its “relatives””
-
Checking the stability and reliability of models: bootstrapping
-
Bayesian probability and statistics
-
How to quickly “find menus” in spss
-
Structural Modeling
-
Publishing analysis results on the web: cdsr
-
Creation, operationalization and delivery of models: spss modeler and watson studio
-
Section results