Survey Design and Sampling Procedures

Trolling for tweets and scraping for sentiment are the rage now, but how reliable are the data?  There is still a role for statistically valid surveys.  Learn the principles of survey design in Anthony Babinec's online course "Survey Design and Sampling Procedures" at Statistics.com. For more details please visit at http://www.statistics.com/surveydesign.

"Survey Design and Sampling Procedures" covers the crafting of survey questions, the design of surveys, and different sampling procedures that are used in practice.  Longstanding basic principles of survey design are covered, and the impact of the trend toward increased respondent resistance is discussed.  This is an introductory course with no prerequisites.

Dr. Babinec is President of AB Analytics and formerly Director of Business Development and Director of Advanced Products Marketing at SPSS, and is an expert in survey design.

Who can take this course:

Anyone who needs to know how to design questionnaires or sample and survey respondents to produce usable results.

Course Program:

Course outline: The course is structured as follows

SESSION 1: The art and craft of asking survey questions

• Crafting questions: the context
• Closed-ended versus open-ended questions
• Levels of measurement

SESSION 2: The art and craft of asking survey questions (continued)

• Measuring attitudes and behaviors
• Factorial designs, conjoint studies
• Special considerations for online surveys

SESSION 3: How to design survey studies

• What is a well-designed survey study?
• Experiments and observational studies
• Different types of designs, and their benefits and concerns

SESSION 4: Sampling

• Populations, samples, and inclusion criteria
• Different sampling methods-how to choose
• Determining sample size
• Acceptable response rate, sources of survey error

You will be able to ask questions and exchange comments with the instructors via a private discussion board throughout the course.   The courses take place online at statistics.com in a series of 4 weekly lessons and assignments, and require about 15 hours/week.  Participate at your own convenience; there are no set times when you must be online. You have the flexibility to work a bit every day, if that is your preference, or concentrate your work in just a couple of days.

For Indian participants statistics.com accepts registration for its courses at special prices in Indian Rupees through its partner, the Center for eLearning and Training (C-eLT), Pune.

For India Registration and pricing, please visit us at www.india.statistics.com.

Email: info@c-elt.com

Call: 020 66009116

Websites:

www.india.statistics.com

www.c-elt.com

Introduction to Resampling Methods

Peter started teaching resampling about 2 decades ago, when the idea of drawing "resamples" from your original sample provoked frequent ridicule.  Try some resampling yourself, and learn what it can and can't do, in Peter Bruce’s online course, "Introduction to Resampling Methods" at statistics.com. For more details please visit at http://www.statistics.com/resampling. 

"Introduction to Resampling Methods" introduces the basic concepts and methods of resampling, including bootstrap procedures and permutation (randomization) tests.  The approach of the course is to teach inference: interval estimation, one-two- and k-sample comparisons, correlation, regression - from a resampling perspective, without complex theory, mathematics or confusing statistical notation.  It is a good lead-in to Michael Chernick's Bootstrap course. 

Who can take this course:

Analysts with data or statistics not suitable for standard analysis (small sample sizes, for example, or non-standard statistics), analysts who have had some statistics and want to deepen their knowledge of statistical inference, statisticians unfamiliar with resampling seeking a basic introduction, instructors interested in the easy-to-understand, non-formula-based resampling approach.

Course Program:

Course outline: The course is structured as follows

SESSION 1: The Resampling Approach to Inference

• Historical perspective
• Hypothesis tests vrs. confidence intervals
• Permutation tests
• The bootstrap
• Virtues of a "naive" do-it-yourself approach
• The 4-step process

• Specify population(s)
• Specify resampling procedure
• Calculate statistic or estimate of interest
• Repeat and keep score

• Working with measured data

SESSION 2: Working with Count Data

• The contingency table
• Choice of test statistics
• Fisher's Exact Test
• Chi-Square Test
• Dose-Response Relationship

SESSION 3: Working with Multivariate Data

• ANOVA
• Correlation
• Regression

You will be able to ask questions and exchange comments with the instructors via a private discussion board throughout the course.   The courses take place online at statistics.com in a series of 3 weekly lessons and assignments, and require about 15 hours/week.  Participate at your own convenience; there are no set times when you must be online. You have the flexibility to work a bit every day, if that is your preference, or concentrate your work in just a couple of days.

For Indian participants statistics.com accepts registration for its courses at special prices in Indian Rupees through its partner, the Center for eLearning and Training (C-eLT), Pune.

For India Registration and pricing, please visit us at www.india.statistics.com.

Email: info@c-elt.com

Call: 020 66009116

Websites:

www.india.statistics.com

www.c-elt.com

Introduction to Bayesian Hierarchical and Multi-level Models

Much research involves sampling from groups or subpopulations at different levels - classes, schools, etc.  Peter Congdon's online course "Introduction to Bayesian Hierarchical and Multi-level Models," teaches you how to incorporate prior information about these groups into Bayesian models. For more details please visit at http://www.statistics.com/bayesian-hierarchical-models.

"Introduction to Bayesian Hierarchical and Multi-level Models" extends the Bayesian modeling framework to cover hierarchical models and to add flexibility to standard Bayesian modeling problems.  Participants will learn how to define three stage hierarchical models and to implement them using Winbugs, in multilevel, meta-analytic and regression applications.  Continuous, count and binary outcomes are covered.  Participants will also learn how to assess goodness-of-fit.

Dr. Peter Congdon is a Research Professor in Quantitative Geography and Health Statistics at Queen Mary University of London.  He is the author of several books and numerous articles in peer-reviewed journals.  His research interests include spatial data analysis, Bayesian statistics, latent variable models and epidemiology.

Who can take this course:

Statistical analysts with some familiarity with Bayesian analysis who want to deepen their skill set in Bayesian modeling.

Course Program:

Course outline: The course is structured as follows

SESSION 1 - Defining Bayesian Hierarchical Models

• Overview of application contexts: meta-analysis to summarise accumulated evidence; comparisons of related units (e.g. "league table comparisons" of exam results, hospital mortality rates, etc); rationale for multi-level models in health, education etc
• Defining Hierarchical Bayesian Models. Three stage models.
• Benefits from "borrowing strength" using Bayesian random effect models.
• Measuring model fit for hierarchical models, and procedures for model checking; effective parameters (and DIC)
• Common conjugate hierarchical models with worked examples

SESSION 2 - Bayesian Hierarchical Models for Meta Analysis

• Modelling the variance/covariance in Bayesian random effects models. Alternative priors for variances. Winbugs implementation of these priors.
• Bayesian meta-analysis and pooled estimates in clinical studies and education
• Different meta-analysis schemes (e.g. beta-binomial, logit-normal for binomial data)

SESSION 3 - Multi-Level and Panel Models

• Multi-level models (2 and 3 level models for continuous, count and binary responses) and Winbugs implementation to include data input structures.
• Simple panel models (random intercept, random slope) from a Bayesian perspective.

SESSION 4 - More on Multilevel Models; Hierarchical Bayesian Regression Models

• Crossed and multivariate and multilevel models
• Overdispersed regression options for count and proportion data including negative binomial and beta-binomial regression

You will be able to ask questions and exchange comments with the instructors via a private discussion board throughout the course.   The courses take place online at statistics.com in a series of 4 weekly lessons and assignments, and require about 15 hours/week.  Participate at your own convenience; there are no set times when you must be online. You have the flexibility to work a bit every day, if that is your preference, or concentrate your work in just a couple of days.

For Indian participants statistics.com accepts registration for its courses at special prices in Indian Rupees through its partner, the Center for eLearning and Training (C-eLT), Pune (www.c-elt.com).

For India Registration and pricing, please visit us at www.india.statistics.com.

Email: info@c-elt.com

Call: 020 66009116

Websites:

www.india.statistics.com

www.c-elt.com

Smoothing with P-splines (Using R)

The linear model is ubiquitous in classical statistics, yet real-life data rarely follow a purely linear pattern.  Smoothing is a commonly-used technique in such cases; "Smoothing with P-splines (Using R)" will be offered online at Statistics.com by Dr. Brian Marx and Dr. Paul Eilers. For more details please visit at http://www.statistics.com/smoothing-p-splines.

Real-life data often do not follow a pattern that is well-described by a single simple function of any sort.  Splines are combinations of different functions that are used to describe and model data differentially in a smooth fashion over different ranges.  P-splines are especially popular (over 500 citations for the instructors' original article in Statistical Science introducing P-splines) because they are widely applicable and effective. In this course, you will learn how to use R software to develop P-splines for data smoothing.  You will be introduced to P-splines via B-splines (basis splines), and learn about the role of difference penalties.  You will learn how to balance the competing demands of fidelity to the data and smoothness, and how to optimize the smoothing.  The final session of the course covers multidimensional smoothing.

Dr. Brian Marx is Professor of Statistics at Louisiana State University, and has taught Categorical Data Analysis for over ten years. He is currently serving as Chair of the Statistical Modelling Society and is the Coordinating Editor of Statistical Modelling: An International Journal. Dr. Marx has numerous publications in peer reviewed journals.

Dr. Paul Eilers is Professor of Genetical Statistics at the Erasmus University Medical Center (Netherlands). Dr. Eilers' research interests include genomic data analysis (esp. high throughput genomic data), chemometrics, smoothing, longitudinal data analysis, survival analysis, and statistical computing.  Dr. Eilers' statistical hobby is filtering and smoothing of time series and signals from chemical instruments.

Who can take this course:

Medical and social science researchers, data miners, environmental analysts;  any researcher who must develop statistical models with "messy" data.

Course Program:

Course outline: The course is structured as follows

SESSION 1:  Smoothing via Regression - Local vs Global Bases

• Global bases can be ineffective
• Local bases are attractive
• B-splines
• Difference penalties

SESSION 2: Introducing P-splines

• Dealing with non-normal data
• Moving from GLM to P-spline
• Density estimation
• Variance smoothing

SESSION 3: Optimizing the Smoothing

• Fidelity to the data vs smooth curve
• Cross-validation, AIC
• Error bands

SESSION 4: Multidimensional Smoothing

• Generalized Addition Models
• Varying coefficient models
• Tensor products

You will be able to ask questions and exchange comments with the instructors via a private discussion board throughout the course.   The courses take place online at statistics.com in a series of 4 weekly lessons and assignments, and require about 15 hours/week.  Participate at your own convenience; there are no set times when you must be online. You have the flexibility to work a bit every day, if that is your preference, or concentrate your work in just a couple of days.

For Indian participants statistics.com accepts registration for its courses at special prices in Indian Rupees through its partner, the Center for eLearning and Training (C-eLT), Pune (www.c-elt.com).

For India Registration and pricing, please visit us at www.india.statistics.com.

Email: info@c-elt.com

Call: 020 66009116

Websites:

www.india.statistics.com

www.c-elt.com

Logistic Regression

You may know the joke:  "There are just 10 types of people - those who understand binary and those who don't." 

Logistic regression (modeling 0/1 data) has become a ubiquitous technique, particularly since the rise of "big data" analytics and businesses' need for classification models (purchase or not, click or not, repay loan or not, etc.). 0/1 (yes/no) data ("same/different," "occur/don't occur," ...) are ubiquitous and modeling them is the province of logistic regression.  Like many important ideas in statistics, logistic regression did not become practical until the advent of computing power and availability.  Learn all about it in Joseph Hilbe's online course, "Logistic Regression“. For more details please visit at http://www.statistics.com/logistic.

"Logistic Regression" will cover the functional form of the logistic model and how to interpret model coefficients the concepts of "odds" and "odds ratio" are examined, as well as "risk ratio" and the difference between the two statistics. Our emphasis is on model construction, interpretation, and goodness of fit. Exercises include hands-on computer problems.

Dr. Joseph Hilbe, the instructor, the author of "Logistic Regression Models," and also "Generalized Linear Models and Extensions" and "Generalized Estimating Equations," was, until recently, the software reviews editor for "The American Statistician."  An ASA Fellow, he is currently Emeritus Professor at the University of Hawaii and Solar System Ambassador with NASA's Jet Propulsion Laboratory at California Institute of Technology.  Participants can ask questions and exchange comments directly with Dr. Hilbe via a private discussion board throughout the period.

As with all online courses at statistics.com, there are no set hours when participants must be online; course work requires about 15 hours per week.

Aim of Course:

Logistic regression is one of the most commonly-used statistical techniques. It is used with data in which there is a binary (success-failure) outcome (response) variable, or where the outcome takes the form of a binomial proportion. Like linear regression, one estimates the relationship between predictor variables and an outcome variable. In logistic regression, however, one estimates the probability that the outcome variable assumes a certain value, rather than estimating the value itself. This course will cover the functional form of the logistic model and how to interpret model coefficients. The concepts of "odds" and "odds ratio" are examined, as well as "risk ratio" and the difference between the two statistics. Our emphasis is on model construction, interpretation, and goodness of fit. Exercises include hands-on computer problems.

Who Should Take This Course:

Medical researchers, epidemiologists, forensic statisticians, environmental scientists, actuaries, data miners, industrial statisticans, sports statisticians, and fisheries, to name a few, will all find this course useful. It is an essential course for anyone who needs to model data with binary or categorical outcomes, and who need to estimate probabilities of given outcomes based on predictor variables.

Course Program:

Course outline: The course is structured as follows

 

SESSION 1: Basic Terminology and Concepts

• Software for modeling logistic regression: Stata, R, SAS, SPSS, other
• History of the logistic model
• Concepts related to logistic regression
• 2x2, 2xn models of odds and risk ratios
• Fitting algorithms

SESSION 2: Logistic Model Construction

• Derivation of the binary logistic model
• Model-building strategies
• Link tests, partial residual plots
• Standard errors: scaling, bootstrap, jackknife, robust
• Interpreting odds ratios as risk ratios - criteria
• Stepwise methods, missing values, constrained coefficients, etc Construction and interpretation of interactions

 

SESSION 3: Analysis, Fit, and Interpretation of the Logistic Model

• Goodness of fit tests
• Information criterion tests
• Residual analysis
• Validation models

SESSION 4: Binomial Logistic Regression and Overdispersion

• The meaning and types of overdispersion
• Simulations: detecting apparent vs real overdispersion
• Methods of handling real overdispersion

You will be able to ask questions and exchange comments with the instructors via a private discussion board throughout the course.   The courses take place online at statistics.com in a series of 4 weekly lessons and assignments, and require about 15 hours/week.  Participate at your own convenience; there are no set times when you must be online. You have the flexibility to work a bit every day, if that is your preference, or concentrate your work in just a couple of days.

For Indian participants statistics.com accepts registration for its courses at special prices in Indian Rupees through its partner, the Center for eLearning and Training (C-eLT), Pune (www.c-elt.com).

For India Registration and pricing, please visit us at www.india.statistics.com.

If you have any query please feel free to call me or write to me. 

For More details contact at

Email: info@c-elt.com

Call: 020 66009116

Websites:

www.india.statistics.com

www.c-elt.com