Introductory Statistics for Laboratorians dealing with High Throughput Data sets

1. Introductory Statistics for Laboratorians dealing with High Throughput Data sets Centers for Disease Control

2. Correlates of Quality of Life in Women with Cervical Cancer • Research Question: • What factors are related to quality of life in women with cervical cancer? • Outcome variable (Y): • WHO Quality of Life Scale completed by each of the women • Predictor variables: • X1: Hours per week spent in coping behaviors (counseling, support group, etc.) • X2: Cancer stress questionnaire score • X3: Spiritual well-being questionnaire score • X4: maladaptive coping questionnaire score • X5: reproductive concerns questionnaire score • X6: social support evaluation score

3. Multiple Regression • Simple Regression • Scatterplot of relationship between two variables • Fit a straight line to the scatterplot • Slope = b1 • Y-intercept = b0 • Use the equation of the line to compute predicted values for unknowns

4. Multiple Regression • Scatterplot of relationship between 3 or more variables (3D or more plot) (cluster of dots in space) • Fit a plane to the cluster of dots in 3 or more dimensions • Get a slope for each predictor variable • Slopes: b1, b2, b3, etc. • Y-intercept: b0 • Equation of the form:

5. Significance Tests • Overall significance: • Null Hypothesis that there is no relationship between any of the X’s and Y • Significance of each slope (b) • Null Hypotheses of the form b1is not significantly different from zero • One test for each b, tells if that particular X variable is contributing to the ability to predict Y.

6. Results • Results : Participants were predominantly married, non-Hispanic White, with a mean age at diagnosis of 37 years and a mean age at interview of 45 years. This disease-free sample enjoys a good QOL, with physical, social, and emotional functioning comparable to or better than comparative norms. However, certain psychological survivorship sequelae and reproductive concerns persist.

7. Participants reporting good QOL were less likely to report ongoing coping efforts related to having had this illness and were more likely to report greater social support, greater sexual pleasure, and less cervical cancer-specific distress. In a multiple-regression model, cancer-specific distress, spiritual well-being, maladaptive coping, and reproductive concerns accounted for 72% of the variance in QOL scores.

8. Fifty-nine percent of respondents expressed that they would likely participate in a counseling program today to discuss psychosocial issues raised by having had cervical cancer, and 69% stated that they would have attended a support group program during the initial treatment if it had been offered.

9. Conclusions • Conclusions : This information provides insight into the complex survivorship relationships between QOL and sequelae of cervical cancer for women diagnosed during childbearing years. Therefore, it is important for health care professionals to recognize that aspects of cancer survivorship continue to require attention and possible follow-up care.