Determination of Confidence Intervals in Non-normal Data: Application of the Bootstrap to Cocaine Concentration in Femoral Blood

Abstract

Calculating the confidence interval is a common procedure in data analysis and is readily obtained from normally distributed populations with the familiar [Formula: see text] formula. However, when working with non-normally distributed data, determining the confidence interval is not as obvious. For this type of data, there are fewer references in the literature, and they are much less accessible. We describe, in simple language, the percentile and bias-corrected and accelerated variations of the bootstrap method to calculate confidence intervals. This method can be applied to a wide variety of parameters (mean, median, slope of a calibration curve, etc.) and is appropriate for normal and non-normal data sets. As a worked example, the confidence interval around the median concentration of cocaine in femoral blood is calculated using bootstrap techniques. The median of the non-toxic concentrations was 46.7 ng/mL with a 95% confidence interval of 23.9-85.8 ng/mL in the non-normally distributed set of 45 postmortem cases. This method should be used to lead to more statistically sound and accurate confidence intervals for non-normally distributed populations, such as reference values of therapeutic and toxic drug concentration, as well as situations of truncated concentration values near the limit of quantification or cutoff of a method.

Social Media and Body Dissatisfaction: Investigating the Attenuating Role of Positive Parent–Adolescent Relationships

“…Predictors were centered prior to computing interaction terms (Cohen et al 2003). Bootstrap confidence intervals were calculated to deal with the slightly skewed distribution of body dissatisfaction (Desharnais et al 2015).…”

Computer-controlled closed-loop drug infusion system for automated hemodynamic resuscitation in endotoxin-induced shock

“…We used the non-parametric Spearman correlation coefficient to examine associations between variables, and interpreted the correlations coefficients using Cohen’s conventions [ 23 ] (.10 small, .30 moderate, .50 large). To determine 99% confidence interval (CI) of the difference between variables, we used a bootstrap technique (1000 replicates) [ 17 , 24 ]. Statistical analyses were performed using commercially available software (Statistica, Statsoft, Inc., Tulsa, OK, USA).…”

Games-based trust: Role-playing the administrative experience of immigrants

“…Whites) how to trust in immigrants (Kuhn, 1996). Departures from normality due to small sample sizes are also addressed through the use of bias-corrected confidence intervals, which confirm significant associations found here (Barber and Thompson, 2000;Desharnais et al, 2015). The self-reported nature of survey items additionally complicates matters, as survey questions are prone to social desirability responses.…”

“…That said, departures from normality become less meaningful as sample sizes increase, and are thought to be robust to violations in samples as small as 30 (Ghasemi and Zahediasl, 2012). Bootstrapping the bias-corrected confidence intervals can further address normality concerns, as a conservative test of significance (Barber and Thompson, 2000; Desharnais et al, 2015). Levene’s test was used to meet homogeneity of variance assumptions across semesters and role-playing groups.…”