Effects Of Additional Variables Research Articles

Effect of Additional Variables on Regression Diagnostics

Effect of Additional Variables on Regression Diagnostics

Reproductive outcomes determine the timing of arrival and settlement of a single-brooded Nearctic–Neotropical migrant songbird (Catharus fuscescens) in South America

ABSTRACT The breeding seasons of Nearctic–Neotropical migratory songbirds are thought to be constrained by the temporal demands of subsequent seasons. In any population, a subset of adults may fail to fledge young despite repeated attempts, which may prolong reproductive effort. We used geolocators to track the single-brooded Veery (Catharus fuscescens) from a Delaware, USA, breeding site to South America and back, to examine how reproductive failure and the timing of reproduction affected the spatiotemporal aspects of the nonbreeding season. We also considered the effect of additional variables: age, sex, time in population (i.e. experience), productivity, and nutrition via ptilochronology. We found that clutch completion date and reproductive success were significant predictors of South American entry and arrival date at first winter sites, respectively. Reproductive success also influenced the latitude of winter sites. For females only, there was a consequence of reproductive failure: unsuccessful fema...

Influences on the intention to enter higher education: the importance of expected returns

Research points to several determinants of the private demand for higher education, which include individual, social, economic, and institutional variables. Of these variables, economic factors are considered to be highly relevant, especially in the current financial crisis. The paper investigates the link between expected rates of return to higher education and the intention of secondary education graduates to pursue university studies. The effect of additional variables influencing the choice of higher education (ability, gender, socioeconomic status [SES], type of secondary school, area of residence) is also examined. The findings show a significant link between the perceived rates of return and the intention of secondary school graduates to enter tertiary education. With the exception of residence, all independent variables are strong determinants of the intention to enter higher education. The paper also points to the considerable effect of the financial crisis on students’ expectations in that respondents reported lower perceived rates of return compared to previous estimates in the same country. The implications of the findings for higher education policy are discussed.

Clinical Outcome in Patients With Peripheral Artery Disease and Renal Artery Stenosis

There is controversy on the influence of renal artery stenosis on outcome in patients with peripheral artery disease. The 12-month impact of renal artery stenosis on declining renal function, control of hypertension, and incidence of major cardiovascular events in 100 consecutive patients undergoing angiography for peripheral artery disease was evaluated. A total of 60 patients had renal artery stenosis: 32 mild, 16 moderate, and 12 severe stenosis. There were no significant differences in either the decline of renal function (2.7 +/- 18% vs 0.9 +/- 16%), control of hypertension (139 +/- 16 vs 139 +/- 22 mm Hg) or number of antihypertensive drugs (1.8 +/- 1.0 vs 1.6 +/- 0.8). Patients with renal artery stenosis had an increased incidence of major cardiovascular events (odds ratio: 2.3; 95% confidence interval: 1.03-5.4), but on multivariate analysis its influence disappeared. Patients with peripheral artery disease having renal artery stenosis had similar decline of renal function and control of hypertension. They had an increased incidence of major cardiovascular events, but it may be explained by the confounding effect of additional variables.

Comparison of experimental assessment methods to estimate the strength of gear teeth

Regression type methods for analysing proportions, known as probit and logit analyses, are compared with the British Standard BS 3518 (part 5) recommendations for analysis of staircase designs. All these methods are found to give similar results. Different experimental strategies, involving fixed sample sizes for load levels chosen in advance, are compared with staircase designs. Recommendations for practice are made. Finally, the versatility of the logit analysis for investigating the effects of additional variables, such as additional treatments during manufacture, is demonstrated through a case study.

Influential Variables in Linear Regression

Based on a statistic proposed by Cook (1977) for the detection of influential observations, we propose a measure for the influence of variables in linear-regression models, and we introduce the leverage of a variable. The new diagnostic can be used to identify important collinearities and, in the checking of a fitted model, to assess the effect of additional variables on the parameter estimates.

The effect of additional variables on factor indeterminacy in models with a single common factor

“Determinate” solutions for the indeterminate common factor ofp variables satisfying the single common factor model are not unique. Therefore an infinite sequence of additional variables that conform jointly with the originalp variables to the original single common factor model does not determine a unique solution for the indeterminate factor of thep variables (although the solution is unique for the factor of the infinite sequence). Other infinite sequences may be found to determine different solutions for the factor of the originalp variables. The paper discusses a number of theorems about the effects of additional variables on factor indeterminacy in a model with a single common factor and draws conclusions from them for factor theory in general.

CIGARETTE CONSUMPTION AND DEATHS FROM CORONARY HEART-DISEASE

There is a positive association between cigarette smoking and coronary heart-disease (C.H.D.). In non-fatal myocardial infarction a dose-response relation persists even after the effects of additional variables have been controlled for. The relation between cigarette consumption and deaths from C.H.D. was investigated in a matched-pair case/control study. The overall simple matched-pair risk ratio (R.R.) between current smokers and non-smokers was 1·9 (95% confidence limits 1·5-2·4). For smokers of fewer than 20 cigarettes per day, the R.R. was 1·2; at a level of 21-40 cigarettes per day the R.R. was 2·3; and for smokers of 41 + cigarettes per day, the R.R. was 4·0. A similar relation was found after adjustment for additional variables. These results suggest that the heaviest smokers could halve their risk of death from C.H.D. by

Some models for individual-group comparisons and group behavior

In problem solving situations, it has been suggested the superiority of groups over individuals is due simply to the fact that groups consist of several individuals. In this paper, the Lorge and Solomon [1955] approach to such situations is reexamined using the method of maximum likelihood. Extensions to trichotomous response situations are also presented, and the resulting models are applied to data gathered by Staub [1970]. The partitioning of the likelihood ratio goodness-of-fit statistic is then discussed in order that the effects of additional variables on the response variate can be assessed. Finally, the small sample behavior of the likelihood ratio statistic is examined.

Precipitation and the levels of Lakes Michigan and Huron

A number of correlations of precipitation and the levels of Lakes Michigan and Huron indicated a computable relationship and also an apparent lag between precipitation and its effect. When land-area precipitation only was considered, a correlation of lake level with four-year precipitation data gave a correlation coefficient of 0.69. A correlation of precipitation and the difference in annual mean levels indicated a correlation coefficient of 0.68. In order to include the effect of additional variables, a correlation of water-year precipitation on a two-year basis and net basin supply was made. A correlation coefficient of 0.79 was obtained, and a lag between precipitation and its effect was again indicated. Monthly precipitation values were correlated with seasonal changes in level, and correlation coefficients of 0.85 and 0.81 were obtained with antecedent precipitation and the seasonal rise and antecedent precipitation and the seasonal fall in levels, respectively. Finally, it is pointed out that geological conditions in the Michigan-Huron basin provide a physical reason for the indicated lag between precipitation and its effect.