China
Africa
Explore chapters and articles related to this topic
Can Objects be Moral Agents? Posthuman Praxis in Public Transportation
Published in Kristen R. Moore, Daniel P. Richards, Posthuman Praxis in Technical Communication, 2018
Meredith A. Johnson, Nathan R. Johnson
Practically, to gather data, we deployed a modified version of post hoc analysis, a methodology that typically uses statistical methods (Ramsey, 2010, p. 1057). Post hoc analyses can be used to identify patterns in data not conceptualized before the research. This often happens when the data collection process yields a potentially valuable yet unexpected finding. Thus, the process of data collection provides the warrant for conducting a post hoc analysis. The value of this type of research while using statistical methods is that it allows for better hypotheses to be formed in the future by locating new theories or explanations. We suggest similar value in for our qualitative approach.
Applied Univariate Statistics
Published in Nick Zacharov, Sensory Evaluation of Sound, 2018
Per Bruun Brockhoff, Federica Belmonte
Generally, a statistical analysis will include the following steps and you may have to carry out several iterations:Exploratory analysis (descriptive statistics). Getting an initial idea of the important structures within the data and clarifying the basic (meta) settings of the data at hand with respect to (independent) variables and scales (dependent variables) used. Simple summary/descriptive statistics, and one- and two-variable graphics such as scatter plots and box plots.Modelling. Formal identification of the overall important structures in the data set through statistical modelling, including hypothesis testing: Noise (random) effects as well as systematic (fixed) effects modelled through the design factors and covariates of the experiment. Answering overall questions such as: Whether synergistic and/or antagonistic effects, also jointly known as interaction effects, play a significant and important role for the information to be extracted?Validation and assumptions. Checking the standard assumptions for the models used, including e.g. assumptions of normality, homogeneity of variance and independence to the extent that they are relevant. Investigating potential outlying and influential observations.Conclusions, summary and final visualisation of results. Summarise the important findings by tables and visualisations. Post hoc analysis in ANOVA settings. Include model estimates, hypothesis test results and uncertainties and confidence intervals from the resulting models.
Life Model for PWM Controlled Induction Motor Insulation using Design of Experiments Method
Published in Electric Power Components and Systems, 2019
Triloksingh G. Arora, Mohan V. Aware
To verify whether significant relation exists between the proposed stress factors and the insulation life, Post hoc test by Bonferroni–Holm method is conducted on the fuzzy logic model results. In the design and analysis of experiments Post hoc analysis consists of investigating into the data after the experiment is concluded. The Post hoc test by Bonferroni–Holm method is a powerful technique to find whether significant relation exists between the factors of a sub-group. Table 4 shows the results of Post hoc test by Bonferroni–Holm method conducted on the fuzzy logic results. This is observed from the test results that the proposed stress parameters significantly affect the life of the insulation. Also between the stress parameters equal variance cannot be assumed as p < 0.05. Therefore, results of fuzzy logic model can be used to obtain life model using Design of Experiments (DoE) method.
Muscle recruitment patterns of the subscapularis, serratus anterior and other shoulder girdle muscles during isokinetic internal and external rotations
Published in Journal of Sports Sciences, 2018
Sylvain Gaudet, Jonathan Tremblay, Mickael Begon
An a priori statistical analysis showed there were no significant differences in muscle activation levels and normalized moments between the swimmers and the physically active adults, therefore all participants were pooled together to increase statistical power. The distribution of the peak EMG and of the normalized peak moment data were examined by use of Q-Q plots, Shapiro-Wilk and Levene tests. A three-factor (2 x 2 x 2) ANOVA model was used to compare the differences in muscle activation and moment production between the different velocities (60–240°/s), direction (internal-external) and type of contraction (concentric-eccentric), for each of the 11 muscles separately. Post hoc analysis was performed to identify where the significant differences occurred using Tukey’s HSD tests.
Effects of firefighting hood design, laundering and doffing on smoke protection, heat stress and wearability
Published in Ergonomics, 2021
Richard M. Kesler, Alex Mayer, Kenneth W. Fent, I-Chen Chen, A. Shawn Deaton, R. Bryan Ormond, Denise L. Smith, Andrea Wilkinson, Steve Kerber, Gavin P. Horn
Descriptive statistics were presented as the mean and standard deviation for each of the survey elements, stratified by hood type and timepoint. Data were analysed using 3 × 2 ANOVA to study the impact of three hood conditions (New-Knit; New-Blocking; Laundered-Blocking) and time (pre, post-activity). Post hoc analysis was conducted using Tukey HSD Tests. All tests corresponding to ANOVA were two-sided at the 0.05 significance level and conducted in SPSS version 23 (IBM, Armonk, NY).