Logistic and multinomial-logit models: A brief review on their modifications and extensions

The work presents various techniques of the logistic and multinomial-logit modeling with their modifications. These methods are useful for regression modeling with a binary or categorical outcome, structuring in regression and clustering, singular value decomposition and principal component analysis with positive loadings, and numerous other applications. Particularly, these models are employed in the discrete choice modeling and the best-worst scaling known in applied psychology and socio-economics studies.

FRIENDSHIP IS SKIN (COLOR) DEEP

Friendships between members of different ethnoracial groups can help to reduce prejudice and ease tensions across ethnoracial groups. A large body of literature has explored possible determinants for the formation of these friendships. One unexplored factor is the role of an individual’s skin color in influencing their opportunities to befriend members of other ethnoracial groups. This study seeks to answer two questions: For ethnoracial minorities, how is an individual’s skin color associated with the likelihood that they will engage in a cross-ethnoracial friendship? Does the role of skin color depend on the ethnoracial combination of the two groups that befriend one another? Using waves 1, 2, and 3 of the National Longitudinal Survey of Freshmen and a series of multinomial logit models, the results suggest that the role of skin color is a function of the relative levels of social status of the two ethnoracial groups that befriend one another. I argue that lighter-skinned members of lower status ethnoracial groups have a greater likelihood of having close friendships with members of higher status ethnoracial groups. There is also limited evidence that darker-skinned members of a higher status group, specifically Asians, have a greater likelihood of having close friends from a lower status group.

The promise and peril of youth entrepreneurship in the Middle East and North Africa

PurposeEntrepreneurship is promoted as a solution to high rates of youth unemployment around the world and especially in the Middle East and North Africa (MENA). This paper investigates the potential for youth entrepreneurship to alleviate unemployment, focusing on Egypt, Jordan and Tunisia.Design/methodology/approachThe authors examine who entrepreneurs are (in comparison to the unemployed), using multinomial logit models. The authors compare entrepreneurs' and wage workers' working conditions and earnings. They exploit panel data to assess earnings and occupational dynamics. They specifically use the Labor Market Panel Surveys of 2012 (Egypt), 2016 (Jordan), and 2014 (Tunisia), along with previous waves.FindingsThe authors find that entrepreneurs are the opposite of the unemployed in MENA. The unemployed are disproportionately young, educated and women. Entrepreneurs are older, less educated and primarily men. Entrepreneurship does not generally lead to higher earnings and does have fewer benefits.Originality/valuePromoting youth entrepreneurship is not only unlikely to be successful in reducing youth unemployment in MENA, but also, if successful, may even be harmful to youth.

relationship between urban form and mode choice in US and Mexican cities: A comparative analysis of workers’ commutes

This paper examines empirical relationships among commuters’ mode choice, metropolitan urban form, and socioeconomic attributes in the 100 largest urban areas in the United States and Mexico. Fitting multinomial logit models to data for more than 5 million commuters and their home urban area, we find several consistent relationships and several important differences in relationships among urban form and travel behavior. In both countries, urban residents living in housing types associated with more centrally located housing in more densely populated urban areas with less roadway are less likely to commute by private vehicle than similar residents in other housing types and other urban areas. In addition to some differences in the strength, significance, and signs of several predictor variables, we find large differences in elasticity estimates across contexts. In particular, the US’s high rates of driving and generally car-friendly urban form mean that even dramatic shifts in urban form or income result in only small predicted changes in the probability of commuting by private vehicle. We conclude that land use and transportation policies likely have a more substantial role in shaping commute patterns in countries like Mexico than in countries like the US.

Uncertain Choices: The Heterogeneous Multinomial Logit Model

In this article, a modeling strategy is proposed that accounts for heterogeneity in nominal responses that is typically ignored when using common multinomial logit models. Heterogeneity can arise from unobserved variance heterogeneity, but it may also represent uncertainty in choosing from alternatives or, more generally, result from varying coefficients determined by effect modifiers. It is demonstrated that the bias in parameter estimation in multinomial logit models can be substantial if heterogeneity is present but ignored. The modeling strategy avoids biased estimates and allows researchers to investigate which variables determine uncertainty in choice behavior. Several applications demonstrate the usefulness of the model.

Did the elderly suffer more from housing deprivation? Evidence from Poland

PurposeThe elderly seem to be more subject to housing problems due to lower income, older age of the housing stock and lower mobility. Nonetheless, housing deprivation (HD) is commonly analysed amongst the general population. Less is known about the differences between age clusters, which seems to be a crucial issue in countries like Poland due to population ageing. What is more, the current literature usually analyses only the occurrence of HD, while also an accumulation of its indicators seems to be substantial. The aim of this article is to identify the differences in HD (its occurrence and accumulation) amongst elderly and non-elderly households and to diagnose the risk factors behind those phenomena.Design/methodology/approachThe HD index was calculated and compared. Next, the multinomial logit models were used to assess risk factors of HD.FindingsThe study showed that, surprisingly, HD in Poland occurs more frequently amongst non-elderly households. The elderly ones suffered more from housing cost overburden, while non-elderly from the overpopulation. In large part, analysed risk factors had a stronger influence on housing conditions of the elderly than non-elderly households.Social implicationsSocial policy tools should focus on the situation of single elderly households, especially living in houses, often in villages. This group is particularly affected by problems with the quality of the dwelling and housing cost overburden.Originality/valueIn the paper, the occurrence and accumulation of HD indicators were analysed. The authors applied a methodological framework that is applicable to other European Union (EU) member states based on the EU Survey on Income and Living Conditions (EU-SILC) data. It is possible to continue the research study and compare different economies.

Sequential Dependencies in Solid and Fluid Intake in Nursing Home Residents with Dementia: A Multistate Model

Nursing home (NH) residents with dementia commonly experience low food intake leading to negative consequences. While multilevel factors influence intake, evidence is lacking on how intake is sequentially associated. This study examined the temporal association between previous and current solid and fluid intake in NH residents with dementia. We analyzed 160 mealtime videos involving 27 residents and 36 staff (53 dyads) in 9 NHs. The dependent variable was the current intake state (fluid, solid, no-intake). Independent variables included the prior intake state, technique of current intake state (resident-initiated, staff-facilitated), duration between previous and current intakes. Covariates included resident and staff characteristics. Two-way interactions of duration and technique with the prior intake state, and resident comorbidity and dementia severity were examined using Multinomial Logit Models. Interactions were significant for technique by comorbidity, technique by dementia severity, technique by prior fluid and solid intake, and duration by prior fluid intake. Successful previous intake increased odds of current solid and fluid intake. Staff-facilitation (vs. resident-initiation) reduced odds of solid and fluid intake for residents with moderately severe (vs. severe) dementia. Higher morbidity decreased odds of solid intake (vs. no-intake) for staff-facilitated intake. Resident with severe dementia had smaller odds of solid and fluid intake for resident-initiated intake. Longer duration increased odds of transition from liquid to solid intake. Findings supported strong sequential dependencies in intake, indicating the promise of intervening behaviorally to modify transitions to successful intake during mealtime. Findings inform the development and implementation of innovative mealtime assistance programs to promote intake.