Issue link: https://www.massageandbodyworkdigital.com/i/68008
SOMATIC RESEARCH Research Terminology ESSENTIAL TERMS ± (plus or minus): Indicates a range of values for results. If massage reduces average headache duration from 8.0 ± 4.3 hours to 4.3 ± 2.3 hours, that means the pre-massage headaches lasted anywhere from 12.3 hours to 3.7 hours, and the post-massage headaches lasted anywhere from 6.6 to 2.0 hours. IMRaD: Mnemonic device for the functional sections of a research article: Introduction, Methods, Results, and Discussion. Mean: A way to calculate an average, derived by adding all the numeric values of the results together, and dividing that sum by the number of results. Null hypothesis: The hypothesis, or proposed explanation, that there is no difference between a treatment group and a non-treatment group in a study. p: The probability that, if the null hypothesis is true, a study would produce particular results through chance alone, and researchers would then mistake a false positive result for a real effect. Power: The probability that a study will detect an effect that is truly present and will not make a false negative error by mistaking the failure to detect a real effect for a true absence of any effect. Expressed as an amount between 0 and 1, meaning from 0% to 100%; for example power= 0.8 means that a real effect was detected 80% of the time. Protocol: A standardized procedure or formula for a treatment during a study so that everyone receiving a treatment receives essentially the same treatment. Sample size: Number of participants (n) in a research study. Statistical significance: The threshold at which the researcher determines that the study has passed the test of not making a false positive error and mistaking an effect due to chance for a real effect, judged by the value of p for the results. Often, but not always, p < 0.05, or detecting an effect that is not really there less than 5% of the time. SUPPLEMENTARY TERMS F(number1, number2): F-value is a ratio between the variation in a group, and variation across groups, associated with performing an analysis of variance, or ANOVA. A high F-value, which often means a value greater than 1.0, indicates a low p, and thus increased confidence in the results being due to a real effect, rather than just chance. Post-hoc analysis: a method of analyzing data and looking for patterns that occurs after ("post") the conclusion of the study, rather than specifying in advance what the researchers will be looking for once the study begins. Repeated measures analysis of variance: A type of analysis of variance (ANOVA), which in general are a set of statistical methods to analyze complicated mixtures of conditions and cause and effect by clustering results into components that are associated with different factors involved. 128 massage & bodywork november/december 2008