Sociology Index

NULL HYPOTHESIS

When testing a research hypothesis, which the researcher has good reason to believe is true, it is customary to use a null hypothesis. Null hypothesis is typically a hypothesis of no difference or of no association between variables.

If the research hypothesis is that men have a higher rate of suicide than do women, the null hypothesis would be that there is no difference in suicide rates between men and women. Researcher then try to disprove the null hypothesis and if they fail to reject it, they accept the research hypothesis.

Large-Scale Simultaneous Hypothesis Testing: The Choice of a Null Hypothesis - Efron, Bradley
Abstract: Current scientific techniques in genomics and image processing routinely produce hypothesis testing problems with hundreds or thousands of cases to consider simultaneously. This poses new difficulties for the statistician, but also opens new opportunities. In particular, it allows empirical estimation of an appropriate null hypothesis.

The empirical null may be considerably more dispersed than the usual theoretical null distribution that would be used for any one case considered separately. An empirical Bayes analysis plan for this situation is developed, using a local version of the false discovery rate to examine the inference issues. Two genomics problems are used as examples to show the importance of correctly choosing the null hypothesis.

The Proposal for a Null Hypothesis Test and Search for New Physics in the Top Dilepton Sample
Andrew Ivanov (University of Rochester), CDF Collaboration 
We propose a null hypothesis test to search for new physics in top dilepton decays. We used data collected by the CDF Run 2 experiment in proton-antiproton collisions at a center-of-mass energy of 1.96 TeV at the Fermilab Tevatron. The test is based on a comparison of several kinematic variables; the Kolmogorov-Smirnov statistic is used to quantify the consistency of the variables' distributions in the data with those expected from the production of t \bart and SM background. In this way we determine an overall confidence level on the hypothesis that the Run 2 data can be explained by purely SM physics.

Statistical Power and the Testing of Null Hypotheses: A Review of Contemporary Management Research and Recommendations for Future Studies - Luke H. Cashen, Louisiana State University, Scott W. Geiger, University of South Florida St. Petersburg 
The purpose of this study is to determine how well contemporary management research fares on the issue of statistical power with regard to studies specifically predicting null relationships between phenomena of interest. This power assessment differs from traditional power studies because it focuses solely on studies that offered and tested null hypotheses. A sample of studies containing hypothesized null relationships was taken from five mainstream management journals over the 1990 to 1999 time period. Results of the power assessment suggest that management researchers’ abilities to affirm null hypotheses are low. On average, the power assessment revealed that for those studies that found nonsignificance of results and consequently affirmed their null hypotheses, the actual Type II error rate was nearly 15 times greater than what is advocated in the literature when failing to reject a false null hypothesis. Recommendations for researchers proposing and testing formal null hypotheses are also discussed.

Sex and Gender Comparisons: Does Null Hypothesis Testing Create a False Dichotomy? 
Olga Eizner Favreau, D�partement de psychologie, Universit� de Montr�al.
In an ongoing debate about the value of doing tests for sex differences, those in favour claim that if sex differences exist, it is important to know about them. However, the null hypothesis (NH) tests that are used for inferring group differences can detect only mean differences and provide no information about how the differences are distributed across groups. Theoretical and empirical examples show how NH rejection can occur when only a small proportion of individuals differ from all others, demonstrating that these tests are incapable of supporting inferences to general group differences.

When is it Acceptable to Accept a Null Hypothesis: No Way, Jose? 
Jose M. Cortina, George Mason University, Robert G. Folger, Tulane University 
Previous research has suggested that there exists a bias in the social sciences against no-effect hypotheses. This is regrettable given the importance of establishing not only when an effect does occur but also the boundary conditions of that effect. The purposes of this article are two-fold The first purpose is to review relevant portions of the history of hypothesis testing in an attempt to identify the sources of bias against hypotheses of no effect.

Random walks in the history of life - James L. Cornette, and Bruce S. Lieberman 
Departments of Geology and Ecology and Evolutionary Biology, University of Kansas, Lawrence; and Department of Mathematics, Iowa State University
The simplest null hypothesis for evolutionary time series is that the observed data follow a random walk. We examined whether aspects of Sepkoski's compilation of marine generic diversity depart from a random walk by using statistical tests from econometrics. Throughout most of the Phanerozoic, the random-walk null hypothesis is not rejected for marine diversity, accumulated origination or accumulated extinction, suggesting that either these variables were correlated with environmental variables that follow a random walk or so many mechanisms were affecting these variables, in different ways, that the resultant trends appear random. The only deviation from this pattern involves rejection of the null hypothesis for roughly the last 75 million years for the diversity and accumulated origination time series.

Rushton's Defenders and Their Hasty Rejection of the Null Hypothesis 
Zack Z. Cernovsky, University of Western Ontario 
Rosenthal and Rubin (1985) pointed out that in research on extreme situations (e.g., new treatments for terninally ill patients) any noticeable statistical trend in the desirable direction is valuable. Their approach is now being misused by those defending Rushton's (1988) "theory" about American Blacks (based on weak trends in excessively suspect data sets). Hasty and eager acceptance of weak, biased, and unrepresentative data as scientific evidence of genetically based and relatively immutable racial differences in human potential amounts to psychological warfare on oppressed racial groups. Similar defamation of vulnerable minorities by Nazi pseudoscientists led to the loss of millions of human lives in the past. Statistical theory classifies similar endeavors as a Type I error (a misleading rejection of the null hypothesis).

Testing the Null Hypothesis of Stationarity Against an Autoregressive Unit Root Alternative - Zhijie Xiao
We propose a new test for the null hypothesis that a time series is stationary around a deterministic trend. The test is valid under general conditions on stationarity. Asymptotic distributions of the test statistic are derived under both the null hypothesis and the alternative hypothesis of a unit root. It is shown that the limiting distribution has the classical Kolmogoroff– Smirnoff form. Critical values for the null distribution are calculated. Consistency of the tests is proved.

Null Hypothesis Significance Testing: Effect Size Matters 
Authors: Gliner J. A.; Vaske J. J.; Morgan G. A.
Abstract: A statistically significant outcome only indicates that it is likely that there is a relationship between variables. It does not describe the extent (strength) of that relationship. In this article, emphasis is placed on the importance of assessing the strength of the relationship between the independent and dependent variables using effect size indices.

A Note on the Bandwidth Choice When the Null Hypothesis is Semiparametric - JORGE BARRIENTOS-MAR�N, Universidad de Alicante - Faculty of Economic and Business Sciences 
Abstract: This work presents a tool for the additivity test. The additive model is widely used for parametric and semiparametric modeling of economic data. The additivity hypothesis is of interest because it is easy to interpret and produces reasonably fast convergence rates for non-parametric estimators. Another advantage of additive models is that they allow attacking the problem of the curse of dimensionality that arises in non- parametric estimation. Hypothesis testing is based in the well-known bootstrap residual process. In nonparametric testing literature, the dominant idea is that bandwidth utilized to produce bootstrap sample should be bigger that bandwidth for estimating model under null hypothesis. 

Reconsidering the null hypothesis: Is maternal rank associated with birth sex ratios in primate groups? - Gillian R. Brown, and Joan B. Silk, Sub-Department of Animal Behaviour, Department of Zoology, University of Cambridge; and Department of Anthropology, University of California. 
Trivers and Willard hypothesized that vertebrates adaptively vary the sex ratio of their offspring in response to the mother's physical condition. This hypothesis has produced considerable debate within evolutionary biology. Here we use meta-analysis techniques to evaluate claims that nonhuman primate females facultatively adjust the sex ratio of their progeny in relation to their own dominance rank in a uniform way.