{[ promptMessage ]}

Bookmark it

{[ promptMessage ]}

# unit4b - Validity:,sampling,or measurementerrors Logic...

This preview shows pages 1–4. Sign up to view the full content.

1 Validity : A valid study has no logic, sampling, or measurement errors. Logic The question that will be asked of the data. Selection or sampling Determine whether to use a sample or a census. Measurement How to measure the characteristic of interest. Variable type: categorical (qualitative) or numeric (quantitative) Why are the data needed? What will the data be used for? What questions are going to be asked of the data? Are the patterns of the past going to be repeated in the future? Census versus sample Nonrandom methods Simple random sampling Stratified sampling Systematic or sequential sampling Cluster or area sampling Sample size Precision - How precise should the measurements be? Accuracy Does the measurement measure what we want it to measure (i.e., say = do)? Reliability Would the measurement be the same if we repeated it?

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
2 Mapping Visual representations of data Histograms and Pareto charts St l t d t l t Stem plots, dot plots Box (and whisker) plots Normal probability plots Length of Hospital Stay Diagnosis Category 10 12 14 8 10 12 ncy 0 2 4 6 8 1-2 3-4 5-6 7-8 9-10 11-12 13-14 15-16 17-18 Length of Hospital Stay (days) Frequency 0 2 4 6 Heart Disease Delivery Pnuemonia Malignant Neoplasms Psychoses Fractures Diagnosis Frequen Length of Hospital Stay Days 18 15 12 9 6 3 Produced with Minitab® Statistical Software Percentage of diabetic Medicare enrollees receiving eye exams among 306 hospital referral regions (2001) Source : Wennberg, J. E. 2005. Data from the Dartmouth Atlas Project Length of Hospital Stay e Probability 1.00 .75 Observed Cumulative Probability 1.00 .75 .50 .25 0.00 Expected Cumulative .50 .25 0.00 Produced with SPSS for Windows Strong Negative Correlation X Y r = -0 86 Strong Positive Correlation X Y Microsoft Excel® screen shots r = -0.86 r = 0.91 Positive Correlation X Y r = 0.70 No Correlation X Y r = 0.06