China
Africa
Explore chapters and articles related to this topic
Estimation and inference
Published in Andrew Metcalfe, David Green, Tony Greenfield, Mahayaudin Mansor, Andrew Smith, Jonathan Tuke, Statistics in Engineering, 2019
Andrew Metcalfe, David Green, Tony Greenfield, Mahayaudin Mansor, Andrew Smith, Jonathan Tuke
McNemar’s test is a paired comparison test for a difference in proportions, and the analysis is based on a normal approximation to the sampling distribution of a single sample proportion. Suppose we have n comparisons of two processes A and B, and the result of each comparison can be classified as: both successful; A successful but B not successful; A not successful but B successful; and both not successful. This classification is shown in Table7.2
Text Mining
Published in Rakesh M. Verma, David J. Marchette, Cybersecurity Analytics, 2019
Rakesh M. Verma, David J. Marchette
This does appear to be slightly better, in fact, than the previous model, although the performance on “ham” is slightly worse. A McNemar test gives a p-value of 0.0001866, so we would be justified in thinking the improvement is significant; however, as mentioned previously, one should be cautious in making this type of comparison. This result is likely to be very specific to this particular corpus of emails, which is unlikely to be representative of emails in general.
Reducing Ergonomic Injuries for Librarians Using a Participatory Approach
Published in Marcelo M. Soares, Franscisco Rebelo, Advances in Usability Evaulation, 2013
The numbers (proportions) of changes in subjects’ overall health rating for the three categories of “improved”, “worsened”, or “no change” were 6 (15%), 10 (26%), and 23 (59%). The χ2 value for the McNemar test was 0.56, which did not show a significant difference between “improved” and “worsened” responses (p = 0.45).
Identifying association between pedestrian safety interventions and street-crossing behavior considering demographics and traffic context
Published in Journal of Transportation Safety & Security, 2020
Sha Mamun, Franklin J. Caraballo, John N. Ivan, Nalini Ravishanker, Rebecca M. Townsend, Yaohua Zhang
The McNemar’s test is a nonparametric method used on paired dichotomous observations to test the significance of the difference between proportions, particularly used on before and after studies in which the effectiveness of a treatment needs to be assessed (Lu, 2010). It was introduced in 1947 by Quinn McNemar and is performed by using 2 × 2 contingency tables of the form shown in Figure 2. The statistic of the McNemar’s test has a chi- squared distribution with one degree of freedom, and the formula is shown in Equation 1, which is the formula used by Statistical Analysis Software (SAS) (Stokes, Davis, & Koch, 2000). where A and D represent, in the case of this study, the number of respondents who changed their behavior from “compliant” to “noncompliant,” and from “noncompliant” to “compliant,” respectively. The outcome of this analysis provides the number of respondents who changed their answers positively or negatively, as well as the ones who stayed the same, giving insight about the effectiveness of the pedestrian safety interventions on their crossing behavior.
Winter wheat mapping using a random forest classifier combined with multi-temporal and multi-sensor data
Published in International Journal of Digital Earth, 2018
Jiantao Liu, Quanlong Feng, Jianhua Gong, Jieping Zhou, Jianming Liang, Yi Li
As the same validation samples were shared among all the methods, McNemar’s test was needed to objectively compare the results. McNemar’s test is a non-parametric test which is dependent on confusion matrices with a 2-by-2 table. McNemar’s test is based on the standardized normal test statistic. The detailed usage of McNemar’s test can be found in Foody (2004). Here, McNemar’s test was performed at the 5% significance level. The results are listed in Table 6.