Ways to make data more meaningful

Slides:



Advertisements
Similar presentations
Copyright 2004 David J. Lilja1 Comparing Two Alternatives Use confidence intervals for Before-and-after comparisons Noncorresponding measurements.
Advertisements

Culture and psychological knowledge: A Recap
Method IntroductionResults Discussion Effects of Plans and Workloads on Academic Performance Mark C. Schroeder University of Nebraska – Lincoln College.
SADC Course in Statistics Comparing Means from Independent Samples (Session 12)
Statistical Analysis SC504/HS927 Spring Term 2008 Week 17 (25th January 2008): Analysing data.
Brown, Suter, and Churchill Basic Marketing Research (8 th Edition) © 2014 CENGAGE Learning Basic Marketing Research Customer Insights and Managerial Action.
Full time and part time employment Coventry population in employment by gender Source: Annual Population Survey, Office for National Statistics
‘How to evaluate your own work’ Dr. Catrin Eames Centre for Mindfulness Research and Practice Workshop for the ‘Mindfulness Now’ conference,
Statistics Used In Special Education
Employment, unemployment and economic activity Coventry working age population by disability status Source: Annual Population Survey, Office for National.
Employment, unemployment and economic activity Coventry working age population by gender Source: Annual Population Survey, Office for National Statistics.
Psychology’s Statistics Statistical Methods. Statistics  The overall purpose of statistics is to make to organize and make data more meaningful.  Ex.
 Frequency Distribution is a statistical technique to explore the underlying patterns of raw data.  Preparing frequency distribution tables, we can.
Source: Annual Population Survey, Office for National Statistics. Full time and part time employment Coventry population.
1 of 65 Inferential Statistics I: The t-test Experimental Methods and Statistics Department of Cognitive Science Michael J. Kalsher.
Lecture 2 Review Probabilities Probability Distributions Normal probability distributions Sampling distributions and estimation.
Chapter 7 Sampling Distributions Statistics for Business (Env) 1.
Chapter One Data Collection 1.2 Observational Studies; Simple Random Sampling.
Employment, unemployment and economic activity Coventry working age population by ethnicity Source: Annual Population Survey, Office for National Statistics.
Evaluating Impacts of MSP Grants Ellen Bobronnikov Hilary Rhodes January 11, 2010 Common Issues and Recommendations.
Statistics Who Spilled Math All Over My Biology?!.
Conducted in the summer of 2012 at 19 COCA-I camps, with a total participation of 2725 campers.
Introduction to statistics I Sophia King Rm. P24 HWB
Applied Opinion Research Training Workshop Day 3.
1 HETEROSCEDASTICITY: WEIGHTED AND LOGARITHMIC REGRESSIONS This sequence presents two methods for dealing with the problem of heteroscedasticity. We will.
Dr. Jeffrey Oescher 27 January 2014 Technical Issues  Two technical issues  Validity  Reliability.
EQUALITY & DIVERSITY UP DATING TRAINING Jan Tothill September 2015.
1 Perspectives on the Achievements of Irish 15-Year-Olds in the OECD PISA Assessment
AP PSYCHOLOGY: UNIT I Introductory Psychology: Statistical Analysis The use of mathematics to organize, summarize and interpret numerical data.
A QUANTITATIVE RESEARCH PROJECT -
Measurement & Data Collection
Statistical Significance
Applying thinking skills to EFL classrooms By Mei-Hui Chen Newcastle University
Step 1: Specify a null hypothesis
Different Types of Data
2a. WHO of RESEARCH Quantitative Research
Statistics.
DESCRIPTIVE STATISTICS
Facet5 Audition Module Facilitator Date Year.
Mathematical Presentation of Data Measures of Dispersion
Change Score Analysis versus ANCOVA in Pretest/Posttest Designs:
Research Brief: Mapping A Strategy to Attract the Politically Engaged Student to East Evergreen University Consultants: Elizabeth Goff Scott Gravitt Kim.
Distribution of the Sample Means
Conclusions Context Long-Term Conditions Questionnaire Results
Elementary Statistics
Unit 7 Today we will look at: Normal distributions
Inferential statistics,
By C. Kohn Waterford Agricultural Sciences
Analyzing Reliability and Validity in Outcomes Assessment Part 1
Module 8 Statistical Reasoning in Everyday Life
Comparing Groups.
Warmup To check the accuracy of a scale, a weight is weighed repeatedly. The scale readings are normally distributed with a standard deviation of
Summary descriptive statistics: means and standard deviations:
Daniela Stan Raicu School of CTI, DePaul University
Psych 231: Research Methods in Psychology
A paired-samples t-test compares the means of two related sets of data to see if they differ statistically. IQ Example We may want to compare the IQ scores.
How To conduct a thesis 1- Define the problem
The MidYIS Test.
Intro to Confidence Intervals Introduction to Inference
Psych 231: Research Methods in Psychology
Keller: Stats for Mgmt & Econ, 7th Ed Data Collection and Sampling
Analyzing Reliability and Validity in Outcomes Assessment
How To conduct a thesis 1- Define the problem
InferentIal StatIstIcs
ID Dr. Natālija Nikrus, Dr. Maija Vikmane, Prof. Oskars Kalējs
Understanding Equivalent Fractions
ANalysis Of VAriance Lecture 1 Sections: 12.1 – 12.2
Using the Rule Normal Quantile Plots
Using the Rule Normal Quantile Plots
Chapter 12 Statistics.
Presentation transcript:

Ways to make data more meaningful Maresa Ness & Barry Tolchard Chief Executive Associate Professor Mosaic Clubhouse London South Bank University

Mosaic Clubhouse An opportunity centre for people living with mental health conditions

Explanation of data Aspirations for the future The bars on the bar charts represent everyone who completed the questionnaire at each time point The statistics only compares paired samples That is the same person over all time points Therefore sometimes the bars may not look different But changes are significant as we are only comparing paired data The first chart shows all people who completed at each time point. The second char shows only those who completed all time points.

Changes in wellbeing Short Warwick-Edinburgh wellbeing scale (SWEMWS)

Changes in Wellbeing Interpretation Paired samples Comparing the sample as a whole we find a significance between the first and last completion of the measure This is less accurate, but likely does represent real change There was no significant improvement in Wellbeing between T1-T2 (p = .11) and T1-T3 (p = .70) Wellbeing may change slowly over time and so a different way of measuring this may be used

Wellbeing more importantly SWEMWS Converted to a continuous measure Compared against national data using a simple tool Add in your mean scores and it compares you against everyone else Measures identify people in five percentiles Lowest 20%; 20-40&, 40-60%, 60-80%, highest 20% Mosaic Clubhouse members T2 T1

Changes in self-esteem Rosenberg Self-Esteem Scale *** * *** p < .001 * p < .05

Changes in self-esteem Interpretation Paired samples   Paired Differences t df p M SD SE-M 95% CI of the Difference Lower Upper Pair 1 T1 – T2 -1.25 3.19 .33 -1.90 -.60 -3.84 95 .000 Pair 2 T1 – T3 -1.75 3.34 .68 -3.16 -.34 -2.57 23 .017 There was a significant improvement in Self-esteem between T1-T2 (p < .001) and T1-T3 (p < .05) Numbers were too small to ascertain a significance between T1-T4

Changes in Aspirations for the future Mosaic Aspirations Scale ** * ** p < .001 * p < .05

Changes in Aspirations for the Future Interpretation Paired samples There was a significant improvement in aspirations between T1-T2 (p < .01) and T1-T3 (p < .05) The bar chart previously does not show the paired samples Therefore even though it looks like there was no change In fact when comparing the same person using the paired statistic, there was a significance This is shown in the analysis of the first and last measures, which was highly significant   Paired Differences t df p M SD SE-M 95% CI of the Difference Lower Upper Pair 1 First - Last -.30 2.60 .06 -.41 -.19 -5.41 2200 .000 Pair 2 T1 – T2 .69 2.32 .24 .22 1.16 2.90 95 .005 Pair 3 T12 – T3 -.25 .44 .09 -.44 -.06 -2.77 23 .011

Changes in Personal Development Mosaic Personal Development Scale

Changes in Aspirations for the Future Interpretation Paired samples   Paired Differences t df p Mean Std. Deviation Std. Error Mean 95% CI of the Difference Lower Upper Pair 1 First - Last -.56 6.66 .14 -.83 -.28 -3.97 2219 .000 Pair 2 T1 – T2 .63 10.10 1.03 -1.42 2.67 .61 95 .55 Pair 3 T1 – T3 -7.00 15.42 3.64 -14.67 .67 -1.93 17 .07 There was no significant improvement in personal development between T1-T2 (p = .55) and T1-T3 (p = .07) However, the analysis of the first and last measures was highly significant

Changes in Social Provision Interpretation The Social Provision measure is broken down into four sub-scales. Therefore there is only a first – last analysis of the Social Provision total score This showed a highly significant change

Changes in Social Provision Subscales On all four sub-scales there were significant improvements at various time points

Recap The tool overall was very reliable and the five scales within the tool were equally reliable when considered alone This tool demonstrates change over time and support the model of change introduced by Mosaic. However, more is required to tease out where the change occurs and under what conditions e.g., those attending a particular unit, certain demographic characteristics and so on There were significant improvements over time on all five parts of the Mosaic questionnaire Some of these were evident from T1 to T2 while others only showed at T3 Where some measures e.g., Aspirations initially appeared to be non significant, using the less robust first-last measures did show significance

Benefits & Challenges Culture change Data base changes Training Language Anxiety Excitement Testimonials Telling a whole story

Next Steps Feed-back loop Collecting more data Publications Working on our own measures

Thank you www.mosaic-clubhouse.org