scholarly journals Analyzing ordinal data with metric models: What could possibly go wrong?

2017 ◽  
Author(s):  
Torrin Liddell ◽  
John K. Kruschke
Keyword(s):  
Ordinal Data ◽  
Ordered Probit ◽  
False Alarms ◽  
Parameter Estimates ◽  
Type I ◽  
Type I Errors ◽  
Psychological Science ◽  
Probit Models ◽  
Ordered Probit Models ◽  
Type Ii Errors

We surveyed all articles in the Journal of Personality and Social Psychology (JPSP), Psychological Science (PS), and the Journal of Experimental Psychology: General (JEP:G) that mentioned the term "Likert," and found that 100% of the articles that analyzed ordinal data did so using a metric model. We present novel evidence that analyzing ordinal data as if they were metric can systematically lead to errors. We demonstrate false alarms (i.e., detecting an effect where none exists, Type~I errors) and failures to detect effects (i.e., loss of power, Type II errors). We demonstrate systematic inversions of effects, for which treating ordinal data as metric indicates the opposite ordering of means than the true ordering of means. We show the same problems --- false alarms, misses, and inversions --- for interactions in factorial designs and for trend analyses in regression. We demonstrate that averaging across multiple ordinal measurements does not solve or even ameliorate these problems. We provide simple graphical explanations of why these mistakes occur. Moreover, we point out that there is no sure-fire way to detect these problems by treating the ordinal values as metric, and instead we advocate use of ordered-probit models (or similar) because they will better describe the data. Finally, although frequentist approaches to some ordered-probit models are available, we use Bayesian methods because of their flexibility in specifying models and their richness and accuracy in providing parameter estimates.

scholarly journals Tropical Cyclones and Climate Change Assessment: Part I: Detection and Attribution

2019 ◽  
Vol 100 (10) ◽  
pp. 1987-2007 ◽  
Author(s):  
Thomas Knutson ◽  
Suzana J. Camargo ◽  
Johnny C. L. Chan ◽  
Kerry Emanuel ◽  
Chang-Hoi Ho ◽  
...  
Keyword(s):  
Climate Change ◽  
Historical Data ◽  
False Alarms ◽  
Type I ◽  
Anthropogenic Influence ◽  
Type Ii ◽  
Type I Errors ◽  
Detection And Attribution ◽  
Climate Change Assessment ◽  
Type Ii Errors

AbstractAn assessment was made of whether detectable changes in tropical cyclone (TC) activity are identifiable in observations and whether any changes can be attributed to anthropogenic climate change. Overall, historical data suggest detectable TC activity changes in some regions associated with TC track changes, while data quality and quantity issues create greater challenges for analyses based on TC intensity and frequency. A number of specific published conclusions (case studies) about possible detectable anthropogenic influence on TCs were assessed using the conventional approach of preferentially avoiding type I errors (i.e., overstating anthropogenic influence or detection). We conclude there is at least low to medium confidence that the observed poleward migration of the latitude of maximum intensity in the western North Pacific is detectable, or highly unusual compared to expected natural variability. Opinion on the author team was divided on whether any observed TC changes demonstrate discernible anthropogenic influence, or whether any other observed changes represent detectable changes. The issue was then reframed by assessing evidence for detectable anthropogenic influence while seeking to reduce the chance of type II errors (i.e., missing or understating anthropogenic influence or detection). For this purpose, we used a much weaker “balance of evidence” criterion for assessment. This leads to a number of more speculative TC detection and/or attribution statements, which we recognize have substantial potential for being false alarms (i.e., overstating anthropogenic influence or detection) but which may be useful for risk assessment. Several examples of these alternative statements, derived using this approach, are presented in the report.

scholarly journals An Analysis of Consumer Preferences for Value-Added Seafood Products Derived from Crawfish

2002 ◽  
Vol 31 (2) ◽  
pp. 157-170 ◽  
Author(s):  
R. Wes Harrison ◽  
Timothy Stringer ◽  
Witoon Prinyawiwatkul
Keyword(s):  
Conjoint Analysis ◽  
Consumer Preferences ◽  
Value Added ◽  
Utility Functions ◽  
Ordered Probit ◽  
Probit Models ◽  
Seafood Products ◽  
Ordered Probit Models ◽  
Old Men

Conjoint analysis is used to evaluate consumer preferences for three consumer-ready products derived from crawfish. Utility functions are estimated using two-limit tobit and ordered probit models. The results show women prefer a baked nugget or popper type product, whereas 35- to 44-year-old men prefer a microwavable nugget or patty type product. The results also show little difference between part-worth estimates or predicted rankings for the tobit and ordered probit models, implying the results are not sensitive to assumptions regarding the ordinal and cardinal nature of respondent preferences.

scholarly journals An Improved Progressive TIN Densification Filtering Method Considering the Density and Standard Variance of Point Clouds

2018 ◽  
Vol 7 (10) ◽  
pp. 409 ◽  
Author(s):  
Youqiang Dong ◽  
Ximin Cui ◽  
Li Zhang ◽  
Haibin Ai
Keyword(s):  
Point Clouds ◽  
High Density ◽  
Type I ◽  
Densification Process ◽  
Type I Errors ◽  
Second Stage ◽  
Standard Variance ◽  
Triangular Irregular Network ◽  
Two Stages ◽  
Type Ii Errors

The progressive TIN (triangular irregular network) densification (PTD) filter algorithm is widely used for filtering point clouds. In the PTD algorithm, the iterative densification parameters become smaller over the entire process of filtering. This leads to the performance—especially the type I errors of the PTD algorithm—being poor for point clouds with high density and standard variance. Hence, an improved PTD filtering algorithm for point clouds with high density and variance is proposed in this paper. This improved PTD method divides the iterative densification process into two stages. In the first stage, the iterative densification process of the PTD algorithm is used, and the two densification parameters become smaller. When the density of points belonging to the TIN is higher than a certain value (in this paper, we define this density as the standard variance intervention density), the iterative densification process moves into the second stage. In the second stage, a new iterative densification strategy based on multi-scales is proposed, and the angle threshold becomes larger. The experimental results show that the improved PTD algorithm can effectively reduce the type I errors and total errors of the DIM point clouds by 7.53% and 4.09%, respectively, compared with the PTD algorithm. Although the type II errors increase slightly in our improved method, the wrongly added objective points have little effect on the accuracy of the generated DSM. In short, our improved PTD method perfects the classical PTD method and offers a better solution for filtering point clouds with high density and standard variance.

scholarly journals Statistical Methods for Banking Sectors to Detect the Eligible Customers for Home Loan

2019 ◽  
Vol 8 (4) ◽  
pp. 1849-1853
Keyword(s):  
Sensitivity Analysis ◽  
Statistical Methods ◽  
Sensitivity And Specificity ◽  
Categorical Variables ◽  
Type I ◽  
Type Ii ◽  
Type I Errors ◽  
Fuzzy Analysis ◽  
Type Ii Errors

Nowadays people are interested to avail loans in banks for their needs, but providing loans to all people is not possible to banks, so they are using some measures to identify eligible customers. To measure the performance of categorical variables sensitivity and specificity are widely used in Medical and tangentially in econometrics, after using some measures also if banks provide the loans to the wrong customers whom might not able to repay the loans, and not providing to customers who can repay will lead to the type I errors and type II errors, to minimize these errors, this study explains one, how to know sensitivity is large or small and second to study the bench marks on forecasting the model by Fuzzy analysis based on fuzzy based weights and it is compared with the sensitivity analysis.

An analysis of motorcycle injury and vehicle damage severity using ordered probit models

2002 ◽  
Vol 33 (4) ◽  
pp. 445-462 ◽  
Author(s):  
Mohammed A. Quddus ◽  
Robert B. Noland ◽  
Hoong Chor Chin
Keyword(s):  
Ordered Probit ◽  
Probit Models ◽  
Ordered Probit Models

Comparing the Variances of Two Dependent Groups

1990 ◽  
Vol 15 (3) ◽  
pp. 237-247 ◽  
Author(s):  
Rand R. Wilcox
Keyword(s):  
Maximum Likelihood ◽  
Low Power ◽  
Maximum Likelihood Estimate ◽  
Type I ◽  
Type Ii ◽  
Type I Errors ◽  
Satisfactory Solution ◽  
Scheffe Test ◽  
Modified Maximum Likelihood ◽  
Type Ii Errors

Let X and Y be dependent random variables with variances σ2x and σ2y. Recently, McCulloch (1987) suggested a modification of the Morgan-Pitman test of Ho: σ2x=σ2y But, as this paper describes, there are situations where McCulloch’s procedure is not robust. A subsample approach, similar to the Box-Scheffe test, is also considered and found to give conservative results, in terms of Type I errors, for all situations considered, but it yields relatively low power. New results on the Sandvik-Olsson procedure are also described, but the procedure is found to be nonrobust in situations not previously considered, and its power can be low relative to the two other techniques considered here. A modification of the Morgan-Pitman test based on the modified maximum likelihood estimate of a correlation is also considered. This last procedure appears to be robust in situations where the Sandvik-Olsson (1982) and McCulloch procedures are robust, and it can have more power than the Sandvik-Olsson. But it too gives unsatisfactory results in certain situations. Thus, in terms of power, McCulloch’s procedure is found to be best, with the advantage of being simple to use. But, it is concluded that, in terms of controlling both Type I and Type II errors, a satisfactory solution does not yet exist.

Sign in / Sign up

Export Citation Format

Share Document