Marissa Gargano Sarrynna Sou Susan Edwards Chris Cummiskey

Slides:



Advertisements
Similar presentations
C82MST Statistical Methods 2 - Lecture 5 1 Overview of Lecture Testing the Null Hypothesis Statistical Power On What Does Power Depend? Measures of Effect.
Advertisements

Confidence Intervals, Effect Size and Power
HSRP 734: Advanced Statistical Methods July 24, 2008.
Analysis of frequency counts with Chi square
Hypothesis Testing Steps of a Statistical Significance Test. 1. Assumptions Type of data, form of population, method of sampling, sample size.
1 SOC 3811 Basic Social Statistics. 2 Reminder  Hand in your Assignment 4.
1 Practicals, Methodology & Statistics II Laura McAvinue School of Psychology Trinity College Dublin.
Treatment Effects: What works for Whom? Spyros Konstantopoulos Michigan State University.
Today Concepts underlying inferential statistics
Impact Evaluation Session VII Sampling and Power Jishnu Das November 2006.
Copyright c 2001 The McGraw-Hill Companies, Inc.1 Chapter 7 Sampling, Significance Levels, and Hypothesis Testing Three scientific traditions critical.
Statistical hypothesis testing – Inferential statistics I.
Research Proposal Methods and Procedures. OBJECTIVES  Recognize component subheadings under Methods and Procedures section.  Identify characteristics.
Chapter 3 Goals After completing this chapter, you should be able to: Describe key data collection methods Know key definitions:  Population vs. Sample.
Hypothesis Testing.
Claims about a Population Mean when σ is Known Objective: test a claim.
Academic Viva POWER and ERROR T R Wilson. Impact Factor Measure reflecting the average number of citations to recent articles published in that journal.
SIMPLE TWO GROUP TESTS Prof Peter T Donnan Prof Peter T Donnan.
Cooperative Learning Statistical Significance and Effect Size By: Jake Eichten and Shorena Dolaberidze.
1 Data Linkage for Educational Research Royal Statistical Society March 19th 2007 Andrew Jenkins and Rosalind Levačić Institute of Education, University.
CHAPTER 17: Tests of Significance: The Basics
Using Weighted Data Donald Miller Population Research Institute 812 Oswald Tower, December 2008.
Educational Research Chapter 13 Inferential Statistics Gay, Mills, and Airasian 10 th Edition.
Chapter 9 Introduction to the t Statistic. 9.1 Review Hypothesis Testing with z-Scores Sample mean (M) estimates (& approximates) population mean (μ)
Chapter Outline Goodness of Fit test Test of Independence.
© 2006 by The McGraw-Hill Companies, Inc. All rights reserved. 1 Chapter 7 Sampling, Significance Levels, and Hypothesis Testing Three scientific traditions.
THE ROLE OF SUBGROUPS IN CLINICAL TRIALS Ralph B. D’Agostino, Sr., PhD Boston University September 13, 2005.
Educational Research Inferential Statistics Chapter th Chapter 12- 8th Gay and Airasian.
Statistical Inferences for Variance Objectives: Learn to compare variance of a sample with variance of a population Learn to compare variance of a sample.
Introduction Social ecological approach to behavior change
QNT 351 genius Expert Success/qnt351geniusdotcom FOR MORE CLASSES VISIT
Introduction Social ecological approach to behavior change
Scott Elliot, SEG Measurement Gerry Bogatz, MarketingWorks
Learning Objectives : After completing this lesson, you should be able to: Describe key data collection methods Know key definitions: Population vs. Sample.
Independent-Samples T-Test
Education Equity Research Initiative: Focus on Equity to Inform Internal Evaluations lessons from Jordan Amy Mulcahy-Dunn, Chris Cummiskey, Simon King,
Research Design and Methods (METHODOLOGY)
Using Stata to Analyze Complex Survey Data
Marissa Gargano Sarrynna Sou Susan Edwards Chris Cummiskey
Intervention Study: Kenya PRIMR Case Regression Analysis
Carina Omoeva, FHI 360 Wael Moussa, FHI 360
Dr. Amjad El-Shanti MD, PMH,Dr PH University of Palestine 2016
Making Connections: guidance on non-exam assessment
Carrie O’Reilly, Ph.D., M.S.N., RN Touro University Nevada
Hypothesis Tests: One Sample
EGR Teacher Training in Kenya
Testing a Claim About a Mean:  Known
March 2017 Susan Edwards, RTI International
SPSS STATISTICAL PACKAGE FOR SOCIAL SCIENCES
R Data Manipulation Bootstrapping
Finding Disadvantaged Schools/Students
Simulation: Sensitivity, Bootstrap, and Power
Chapter 6 Making Sense of Statistical Significance: Decision Errors, Effect Size and Statistical Power Part 1: Sept. 18, 2014.
Chapter 12: Comparing Independent Means
Hypothesis Tests for Proportions
Categorical Data Analysis Review for Final
Chapter 1 The Where, Why, and How of Data Collection
Chapter 1 The Where, Why, and How of Data Collection
Power.
Sampling and Power Slides by Jishnu Das.
Analytics – Statistical Approaches
Inferential Statistics & Test of Significance
TG EHIS January 2012 Item 4.1 of the agenda EHIS wave 2 Implementing Regulation Bart De Norre, Eurostat.
Analyzing Reliability and Validity in Outcomes Assessment
Research Design and Methods
Andrew Jenkins and Rosalind Levačić
The Where, Why, and How of Data Collection
Statistical Power.
Chapter 1 The Where, Why, and How of Data Collection
Presentation transcript:

Marissa Gargano Sarrynna Sou Susan Edwards Chris Cummiskey Complex Data Analysis Improve your data analytic abilities using Stata and Early Grade Reading and Mathematics Assessment Data Sunday March 5, 8:30 – 14:45 Georgia 2 (South Tower) CIES 2017 Downtown Sheraton Atlanta, GA Marissa Gargano Sarrynna Sou Susan Edwards Chris Cummiskey Marissa Gargano Sarrynna Sou Susan Edwards Chris Cummiskey

Outline of Intervention Workshop Introduce the PRIMR Study Background Establish and address direct aims Difference-in-Difference (DiD) Values Explain what a DiD is and how it’s calculated Regressions Set up models; Link DiD and t-tests calculations to the regression output

Overview of PRIMR Kenya Study Scope Apply innovative, data-based instructional improvement methods to increase students’ fundamental skills in reading and mathematics Purpose Create a sustainable reading and mathematics program could be implemented Assessment tool EGRA-EGMA Time frame January 2012 – October 2013 Grades 1 and 2

Overview of Workshop Data Set Subset of PRIMR Kenya study C1 and C3 Oct 2012 – 2013 Grade 2 Midterm A will be our baseline Disclaimer: Results from this workshop will not reflect published estimates from this study

Overview of Workshop Data Set Treatment/ year 2012 2013 Total Control 377 400 777 Full Treatment 536 522 1,058 913 922 1,835 Gender/ Grade Second Male 893 Female 942 1,835

Aims Primary AIM Did PRIMR have an effect on pupil achievement in reading and math? “Did the treatment work?” Secondary AIM Was the treatment more effective in the public or non-formal schools? Did the treatment work?

Sample How were the students sampled? Do we resample at end line or use the same schools from baseline? Intervention Assessment Only? Resample not needed. Another population level estimate? Depends . . . Is there concern about fidelity of implementation? Assumption: No time changes in the population.

Baseline Sampling Variable is nonformal

10 minute break

Appending and Merging Goal: Set up the final data set with baseline and endline data. Problem: We have two separate data sets Baseline Endline How do we bring these data sets together in Stata?

Appending and Merging - WHEN Stacking Datasets – one long dataset when you have separate baseline, midterm, or end line data Merge Linking Datasets – one wide dataset Same students are tested in two different languages School Level Demographics (SSME) Longitudinal Set Up Data

Appending and Merging – APPENDING IN STATA Example Code:

Address Aims Goal: Address the following aims by calculating simple baseline and end line comparisons. Did PRIMR have an effect on pupil achievement in reading and math? Was the treatment more effective in the public or non-formal schools? What do we want to compare? How do we compare these different groups? Control/ Treatment Gender School type Rural/ urban

Simple Comparisons Null Hypothesis Alternative Hypothesis No difference between the two groups in question Alternative Hypothesis There is a significant difference between the two groups in question Rejection Level (Alpha Level) The probability of the difference not occurring due to chance Test Statistic Level of Significance (p-value) Null Hypothesis- there is no significant difference between the two groups in question Alternative Hypothesis- there is significant difference due to an effect Rejection Level (Alpha Level)- .01, .05, .10 Test Statistic- t –test in this case- used in the hypothesis test to determine significance Level of Significance (p-value)- what is the p value that the t test yields?

Simple Comparisons Activity Please open up do file \\Kenya PRIMR Comparison Activity.do There is a list of questions to help guide you through calculating weighted simple comparisons SVY comment

Example in Do File

Example in Do File Look at the differences!!!!!!!!!

Discussion What did you find most interesting from these comparisons? Were there issues you came across during the activity? If so, may you state them? We see that there a differences between mean values, now how do we measure how big these differences are?

10 Minute Break?