umu.sePublications
Change search
Refine search result
34567 251 - 300 of 303
CiteExportLink to result list
Permanent link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Rows per page
  • 5
  • 10
  • 20
  • 50
  • 100
  • 250
Sort
  • Standard (Relevance)
  • Author A-Ö
  • Author Ö-A
  • Title A-Ö
  • Title Ö-A
  • Publication type A-Ö
  • Publication type Ö-A
  • Issued (Oldest first)
  • Issued (Newest first)
  • Created (Oldest first)
  • Created (Newest first)
  • Last updated (Oldest first)
  • Last updated (Newest first)
  • Disputation date (earliest first)
  • Disputation date (latest first)
  • Standard (Relevance)
  • Author A-Ö
  • Author Ö-A
  • Title A-Ö
  • Title Ö-A
  • Publication type A-Ö
  • Publication type Ö-A
  • Issued (Oldest first)
  • Issued (Newest first)
  • Created (Oldest first)
  • Created (Newest first)
  • Last updated (Oldest first)
  • Last updated (Newest first)
  • Disputation date (earliest first)
  • Disputation date (latest first)
Select
The maximal number of hits you can export is 250. When you want to export more records please use the Create feeds function.
  • 251.
    Södergren, Anna
    et al.
    Umeå University, Faculty of Medicine, Public Health and Clinical Medicine, Reumatology.
    Stegmayr, Birgitta
    Umeå University, Faculty of Medicine, Public Health and Clinical Medicine, Medicine.
    Lundberg, Vivan
    Department of Medicine, Kalix Hospital, Kalix, Sweden .
    Öhman, Marie-Louise
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Wållberg-Jonsson, Solveig
    Umeå University, Faculty of Medicine, Public Health and Clinical Medicine, Reumatology.
    Increased incidence of and impaired prognosis after acute myocardial infarction among patients with seropositive rheumatoid arthritis.2007In: Annals of the Rheumatic Diseases, ISSN 0003-4967, E-ISSN 1468-2060, ISSN Online ISSN 1468-2060, Vol. 66, no 2, p. 263-266Article in journal (Refereed)
    Abstract [en]

    Objective: To examine the incidence and outcome of acute myocardial infarction (AMI) in patients with rheumatoid arthritis compared with the general population, and to examine whether care and treatment of an AMI differs between patients and controls.

    Methods: The Multinational Monitoring of Trends and Determinants of Cardiovascular Disease register for northern Sweden was used to compare those incidences of AMI in a cohort of patients with rheumatoid arthritis with that in the general population. 35 patients with rheumatoid arthritis who had also experienced an AMI were identified. For each patient with rheumatoid arthritis, three controls with a history of AMI but without rheumatoid arthritis were randomly selected from the same register, and matched for age, sex and year of the AMI for evaluation of case fatality and potential differences in treatment of AMI.

    Results: The standardised incidence ratio for AMI was 2.9 in patients with rheumatoid arthritis compared with the general population (p<0.05). During the first 10 years after an AMI, patients with rheumatoid arthritis had a higher overall case fatality compared with controls (hazard ratio (HR) 1.67, 95% confidence interval (CI) 1.02 to 2.71). Survival time was decreased in the rheumatoid arthritis group compared with controls despite the same care and treatment.

    Conclusion: Both the incidence of and case fatality after an AMI were higher among patients with rheumatoid arthritis than among the general population. The results emphasise the necessity of optimising the preventive, diagnostic and caring strategies for AMI in rheumatoid arthritis.

  • 252.
    Södergren, Anna
    et al.
    Umeå University, Faculty of Medicine, Department of Public Health and Clinical Medicine, Reumatology.
    Stegmayr, Birgitta
    Umeå University, Faculty of Medicine, Department of Public Health and Clinical Medicine, Medicine.
    Öhman, Marie-Louise
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Wållberg-Jonsson, Solveig
    Umeå University, Faculty of Medicine, Department of Public Health and Clinical Medicine, Reumatology.
    Increased incidence of stroke and impaired prognosis after stroke among patients with seropositive rheumatoid arthritis2009In: Clinical and Experimental Rheumatology, ISSN 0392-856X, E-ISSN 1593-098X, Vol. 27, no 4, p. 641-644Article in journal (Refereed)
    Abstract [en]

    OBJECTIVE: To examine the incidence of, and outcome after, a stroke in patients with rheumatoid arthritis (RA) compared with the general population. METHODS: The northern Sweden MONICA register was used to compare the incidence of stroke in a cohort of RA patients with the general population. Forty RA patients who had also suffered a stroke were identified. For each patient with RA, three controls with a history of stroke but without RA were randomly collected from the same register, and matched for age and sex. RESULTS: The standardised incidence ratio (SIR) for stroke was 2.7 in RA patients compared with the general population (p<0.05). During the follow-up, RA patients had a higher overall case fatality (CF) following stroke compared with controls (hazard ratio (HR) =1.70, p<0.05). CONCLUSIONS: Both the incidence of a stroke, and the subsequent CF, were higher among RA patients compared with the general population. The results emphasize the necessity of optimising the prevention of stroke and follow-up care after a stroke in RA.

  • 253.
    Tano Bask, Andreas
    et al.
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Jaurin, Johan
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Det elliptiska säkerhetsområdets robusthet: hur robust är metoden med de elliptiska säkerhetsområdena förett symmetriskt men icke normalfördelat datamaterial?2010Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesis
    Abstract [en]

    Quality Control is a term often used within production and is referring to managing processes so they produce capable products. Within Quality Control, process capability index is a common measure to oversee processes. Safety Region Plots were introduced to do this graphically. In Albing & Vännman (2010) the concept of Safety Region Plots is expanded to incorporate an elliptical shape. The method of Elliptical Safety Region Plots assumes a normally distributed data. In this paper we are looking at the robustness of the Elliptical Safety Region Plots if we can assume a symmetrically, but non-normal, distribution. In the results we can conclude that an adjustment is required for symmetric, but non-normal, data if the method in Albing & Vännman (2010) is going to be used. An eventual adjustment is discussed in discussions. To easily be able to use the Elliptical Safety Region Plots mentioned in Albing & Vännman (2010) we have developed a program in RExcel.

  • 254.
    Törnkvist, Birgitta
    Umeå University, Faculty of Social Sciences, Statistics.
    Quantifying Structural Change: A Model Based Approach1988Doctoral thesis, monograph (Other academic)
  • 255.
    Törnkvist, Birgitta
    et al.
    Umeå University, Faculty of Social Sciences, Statistics.
    Henriksson, Widar
    Educational Measurement.
    Repeated Test Taking: Differences Between Social Groups2004Report (Other academic)
    Abstract [en]

    The Swedish Scholastic Aptitude Test (SweSAT) is used in the process of selection to higher education in Sweden. The SweSAT is administered twice a year and consists of 122 items. The total raw score is transformed into a normed score, which is used in the selection procedure. The main purpose of this study was to investigate the effects of repeated test taking regarding the SweSAT. The effects are described in relation to the test taker who chooses to retest. Another purpose was to relate the effects of repeated test taking to the individual subtests of the SweSAT and to analyse the self selection process related to repeated test taking. Multivariate linear models for repeated measurements were used to describe the effects of repeated test taking in relation to the test taker. Calibrated scores were used as a basis for relating the effects to subtests. The summarised conclusions were that test takers with high SweSAT score repeated the test more often than those with low scores. Males and test takers from social group I repeated the test more often than females and social group III, respectively. Repeated test taking had its main effect from the first to the second testing for all social groups and both sexes. This main effect of repeated test taking was related to two subtests, WORD and DTM, for all social groups. For the other subtests, social groups II and III were favoured by subtest DS, social group II was favoured by READ and social group III was favoured by subtest ERC. With reference to total normed test score the results indicated that there was a marginal difference between social groups I and II, when controlled for sex and age. Social group I had a higher mean score than the other social groups. Males had higher mean scores than females. The summarised conclusion was that the differences in mean scores between social groups I and III were higher than between males and females.

  • 256.
    Törnkvist, Birgitta
    et al.
    Umeå University, Faculty of Social Sciences, Statistics.
    Henriksson, Widar
    Educational Measurement.
    SweSAT REPEAT2004Report (Other academic)
    Abstract [en]

    The Swedish Scholastic Aptitude Test (SweSAT) is, together with

    grade point average (GPA), used in the selection procedure for higher education in Sweden. The SweSAT is administered twice a year, it consists of 122 items and the total raw score is transformed to a normed score which is used in the selection procedure. Certain rules apply to the SweSAT when it is used for selection to higher education and one of these rules is that the test taker can repeat the test. The main purpose of this study was to investigate the effects of repeated test taking for the SweSAT. The effects are described in terms of normed score, partly in relation to the rules for selection, and partly in relation to the test taker who elect to retest. The result indicates that the effects of repeated test taking mainly occur between the first and second testing. This is a tendency that is observed when the effects of repeated test taking are analysed in relation to the rules for selection as well as in relation to the test taker. According to the rules for the SweSAT in the selection system the best score is used if a certain applicant has more than one valid score and the model for description of score gains in this study is based on this fact. Another purpose was to relate the effects of repeated test taking to subtests of the SweSAT. Proportional stratified sampling and multivariate linear models are used when describing the effects of repeated test taking in relation to the test taker, while a reference population and a calibrated raw score is used as a basis when the effects are related to subtests. With regard to subtests, the main effect of repeated test taking is related to three subtests: WORD, DS and ERC. The summarised conclusion that is made in this study is that the effects of repeated test taking between the first and second test occasion, in terms of standard deviation units for total normed score for the SweSAT, is about 0.2 and this finding is in congruence with findings in earlier studies. Plausible hypotheses for

    the obtained results are discussed.

  • 257.
    Törnkvist, Birgitta
    et al.
    Umeå University, Faculty of Social Sciences, Statistics.
    Henriksson, Widar
    Educational Measurement.
    Validity issues concerning repeated test taking of the SweSAT2006Report (Other academic)
    Abstract [en]

    The main purpose of this study is to integrate and discuss results from studies that focus on the effects of repeated test taking of the Swe-SAT. Messick’s four-faceted model of validity is used as an integrating and an analytic tool. Another purpose is to use results from supplementary education as a reference in this integration.

    The summarized conclusion is that the existing rule concerning repeated test taking for the Swedish Scholastic Assessment Test (Swe-SAT) is relevant. It is relevant as the test taker has a chance to obtain a good estimate of his or her knowledge and ability. The unintended social consequences are reduced if actions are taken to motivate test takers to repeat the test.

    The conclusion is quite the opposite for supplementary education. A

    change in the rule that eliminates the usage of supplementary grades in the selection procedure will result in higher validity.

    The conclusion is also that the Messick model is a very useful tool for validation. When applying the Messick model, the aim and the direction of the process of validation will be systemized and optimized as well as nuanced.

  • 258.
    van der Linden, Wim J.
    et al.
    CTB/McGraw-Hill, Monterey, California.
    Wiberg, Marie
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Local observed-score equating with anchor-test designs2010In: Applied psychological measurement, ISSN 0146-6216, E-ISSN 1552-3497, Vol. 34, no 8, p. 620-640Article in journal (Refereed)
    Abstract [en]

    For traditional methods of observed-score equating with anchor-test designs, such as chain and poststratification equating, it is difficult to satisfy the criteria of equity and population invariance. Their equatings are therefore likely to be biased. The bias in these methods was evaluated against a simple local equating method in which the anchor-test score was used as a proxy of the proficiency measured by the test and the equating was conditional on this score. The results showed substantial bias for the two traditional methods under a variety of conditions but much smaller bias for the local method. In addition, unlike the traditional methods, the local method appeared to be quite robust with respect to changes in the difficulty and accuracy of the two tests that were equated. But like these methods, it appeared to be sensitive to a decrease in the accuracy of the anchor test as a proxy of the ability measured by the tests.

  • 259.
    Vanhatalo, Erik
    et al.
    Quality Technology, Luleå University of Technology, Luleå, Sweden.
    Kvarnström, Björn
    Quality Technology, Luleå University of Technology, Luleå, Sweden.
    Bergquist, Bjarne
    Quality Technology, Luleå University of Technology, Luleå, Sweden.
    Vännman, Kerstin
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    A method to determine transition time for experiments in dynamic processes2011In: Quality Engineering, ISSN 0898-2112, E-ISSN 1532-4222, Vol. 23, no 1, p. 30-45Article in journal (Refereed)
    Abstract [en]

    Process dynamics is an important consideration during the planning phase of designed experiments in dynamic processes. After changes of experimental factors, dynamic processes undergo a transition time before reaching a new steady state. To minimize experimental time and reduce costs and for experimental design and analysis, knowledge about this transition time is important. In this article, we propose a method to analyze process dynamics and estimate the transition time by combining principal component analysis and transfer function–noise modeling or intervention analysis. We illustrate the method by estimating transition times for a planned experiment in an experimental blast furnace.

  • 260.
    Vonkavaara, Malin
    et al.
    Umeå University, Faculty of Medicine, Department of Clinical Microbiology, Clinical Bacteriology.
    Telepnev, Maxim V
    Umeå University, Faculty of Medicine, Department of Clinical Microbiology, Clinical Bacteriology.
    Rydén, Patrik
    Umeå University, Faculty of Science and Technology, Department of Mathematics and Mathematical Statistics. Umeå University, Faculty of Social Sciences, Department of Statistics. Umeå University, Faculty of Medicine, Department of Clinical Microbiology, Clinical Bacteriology.
    Sjöstedt, Anders
    Umeå University, Faculty of Medicine, Department of Clinical Microbiology, Clinical Bacteriology.
    Stöven, Svenja
    Umeå University, Faculty of Medicine, Department of Clinical Microbiology, Clinical Bacteriology.
    Drosophila melanogaster as a model for elucidating the pathogenicity of Francisella tularensis2008In: Cellular Microbiology, ISSN 1462-5814, E-ISSN 1462-5822, Vol. 10, no 6, p. 1327-1338Article in journal (Refereed)
    Abstract [en]

    Drosophila melanogaster is a widely used model organism for research on innate immunity and serves as an experimental model for infectious diseases. The aetiological agent of the zoonotic disease tularaemia, Francisella tularensis, can be transmitted by ticks and mosquitoes and Drosophila might be a useful, genetically amenable model host to elucidate the interactions between the bacterium and its arthropod vectors. We found that the live vaccine strain of F. tularensis was phagocytosed by Drosophila and multiplied in fly haemocytes in vitro and in vivo. Bacteria injected into flies resided both inside haemocytes and extracellularly in the open circulatory system. A continuous activation of the humoral immune response, i.e. production of antimicrobial peptides under control of the imd/Relish signalling pathway, was observed and it may have contributed to the relative resistance to F. tularensis as flies defective in the imd/Relish pathway died rapidly. Importantly, bacterial strains deficient for genes of the F. tularensis intracellular growth locus or the macrophage growth locus were attenuated in D. melanogaster. Our results demonstrate that D. melanogaster is a suitable model for the analysis of interactions between F. tularensis and its arthropod hosts and that it can also be used to identify F. tularensis virulence factors relevant for mammalian hosts.

  • 261.
    Waara, Rolf
    et al.
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Westin, Kerstin
    Umeå University, Faculty of Social Sciences, Department of Social and Economic Geography.
    Taxis verksamhet och omgivning1991Report (Other academic)
  • 262.
    Waernbaum, I
    et al.
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Blohmé, G
    Ostman, J
    Sundkvist, G
    Eriksson, Jan
    Umeå University, Faculty of Medicine, Department of Public Health and Clinical Medicine, Medicine.
    Arnqvist, H J
    Bolinder, J
    Nyström, Lennarth
    Umeå University, Faculty of Medicine, Department of Public Health and Clinical Medicine, Epidemiology and Public Health Sciences.
    Excess mortality in incident cases of diabetes mellitus aged 15 to 34 years at diagnosis: a population-based study (DISS) in Sweden.2006In: Diabetologia, ISSN 0012-186X, E-ISSN 1432-0428, Vol. 49, no 4, p. 653-659Article in journal (Refereed)
  • 263.
    Waernbaum, Ingeborg
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Covariate selection and propensity score specification in causal inference2008Doctoral thesis, comprehensive summary (Other academic)
    Abstract [en]

    This thesis makes contributions to the statistical research field of causal inference in observational studies. The results obtained are directly applicable in many scientific fields where effects of treatments are investigated and yet controlled experiments are difficult or impossible to implement.

    In the first paper we define a partially specified directed acyclic graph (DAG) describing the independence structure of the variables under study. Using the DAG we show that given that unconfoundedness holds we can use the observed data to select minimal sets of covariates to control for. General covariate selection algorithms are proposed to target the defined minimal subsets.

    The results of the first paper are generalized in Paper II to include the presence of unobserved covariates. Morevoer, the identification assumptions from the first paper are relaxed.

    To implement the covariate selection without parametric assumptions we propose in the third paper the use of a model-free variable selection method from the framework of sufficient dimension reduction. By simulation the performance of the proposed selection methods are investigated. Additionally, we study finite sample properties of treatment effect estimators based on the selected covariate sets.

    In paper IV we investigate misspecifications of parametric models of a scalar summary of the covariates, the propensity score. Motivated by common model specification strategies we describe misspecifications of parametric models for which unbiased estimators of the treatment effect are available. Consequences of the misspecification for the efficiency of treatment effect estimators are also studied.

  • 264.
    Waernbaum, Ingeborg
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Model misspecification and robustness in causal inference: comparing matching with doubly robust estimation2012In: Statistics in Medicine, ISSN 0277-6715, E-ISSN 1097-0258, Vol. 31, no 15, p. 1572-1581Article in journal (Refereed)
    Abstract [en]

    In this paper we compare the robustness properties of a matching estimator with a doubly robust estimator. We describe the robustness properties of matching and subclassification estimators by showing how misspecification of the propensity score model cam result in consistent estimation of the average causal effect. The propensity scores are covariate scores, which are a class of functions that removes bias due to all observed covariates. When matching on a parametric model (e.g. a propensity or prognostic score), the matching estimator is robust to model misspecifications if the misspecified model belongs to the class of covariate scores. The implication is that there are multiple possibilities for the matching estimator in contrast to the doubly robust estimator in which the researcher has two chances to make reliable inference. In simulations, we compare the finite sample properties of the matching estimator with a simple inverse probability weighting estimator and a doubly robust estimator. For the misspecifications in our study the mean square error of the matching estimator is smaller than the mean square error of both the simple inverse probability weighting estimator and the doubly robust estimator-

  • 265.
    Waernbaum, Ingeborg
    Umeå University, Faculty of Social Sciences, Statistics.
    Model-free variable selection in causal inferenceManuscript (Other academic)
  • 266.
    Waernbaum, Ingeborg
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Propensity score model specification for estimation of average treatment effects2010In: Journal of Statistical Planning and Inference, ISSN 0378-3758, E-ISSN 1873-1171, Vol. 140, no 7, p. 1948-1956Article in journal (Refereed)
    Abstract [en]

    Treatment effect estimators that utilize the propensity score as a balancing score, e.g., matching and blocking estimators are robust to misspecifications of the propensity score model when the misspecification is a balancing score. Such misspecifications arise from using the balancing property of the propensity score in the specification procedure. Here, we study misspecifications of a parametric propensity score model written as a linear predictor in a strictly monotonic function, e.g. a generalized linear model representation. Under mild assumptions we show that for misspecifications, such as not adding enough higher order terms or choosing the wrong link function, the true propensity score is a function of the misspecified model. Hence, the latter does not bring bias to the treatment effect estimator. It is also shown that a misspecification of the propensity score does not necessarily lead to less efficient estimation of the treatment effect. The results of the paper are highlighted in simulations where different misspecifications are studied.

  • 267.
    Wang, Qun
    et al.
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Ali-Reza, Rezaie
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Prognos av framtida inkomster hos CSN:s låntagare: En jämförande regressionsanalys mellan flera regressionsmetoder vid vänstertrunkerad data2011Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesis
  • 268.
    Wennberg, Jessica
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Möjligheternas kommun?: En analys av den framtida bostadsförsörjningen i Pajala Kommun2010Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesis
  • 269.
    Westlund, Anders
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Estimation and Prediction in Interdependent Systems in the Presence of Specification Errors1975Doctoral thesis, monograph (Other academic)
    Abstract [en]

    Mult i relät ional models involving simultaneously dependent linear equations are often employed in prediction and regulation of economic systems. The prediction purposes focus their attention on the reduced form of the model, relating each endogenous variable to exogenous variables, predicted outside the system. Different principles for estimating the reduced form parameters have been suggested. These concentrate to a different extent upon sample information and structural a priori information. Knowledge of the properties <bf predictions based on different principles of reduced form estimation is essential when choosing between competing estimation techniques. Reduced form estimation is often subject to many potential sources of specification error. In this work different types of specification errors, essentially due to incorrect omission of exogenous and endogenous variables, are studied in the context of reduced form estimation. A strategy of estimating reduced form parameters in the case of structural forms characterized by heterogenous specification quality is proposed and evaluated analytically as well as by a Monte Carlo study. The Monte Carlo study supports the conjecture that this strategy is better (when considering prediction quality measured by Ball's Q ) than purely direct (unrestricted) or indirect (restricted) reduced form estimation.

  • 270.
    Wiberg, Marie
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    An optimal design approach to criterion-referenced computerized testing2003In: Journal of educational and behavioral statistics, ISSN 1076-9986, E-ISSN 1935-1054, Vol. 28, no 2, p. 97-110Article in journal (Refereed)
    Abstract [en]

    A criterion-referenced computerized test is expressed as a statistical hypothesis problem. This admits that it can be studied by using the theory of optimal design. The power function of the statistical test is used as a criterion function when designing the test. A formal proof is provided showing that all items should have the same item characteristics, i.e. items that have high discrimination, low guessing and difficulty near the cut-off score give the most powerful statistical test. An efficiency study shows how many times more items are needed if nonoptimal items are used instead of optimal items in order to get the same power in the test.

  • 271.
    Wiberg, Marie
    Umeå University, Faculty of Social Sciences, Department of Statistics. Umeå University, Faculty of Social Sciences, Department of Educational Measurement.
    An optimal design approach to criterion-referenced computerized testing2003In: Journal of educational and behavioral statistics, ISSN 1076-9986, E-ISSN 1935-1054, Vol. 28, no 2, p. 97-110Article in journal (Refereed)
    Abstract [en]

    A criterion-referenced computerized test is expressed as a statistical hypothesis problem. This admits that it can be studied by using the theory of optimal design. The power function of the statistical test is used as a criterion function when designing the test. A formal proof is provided showing that all items should have the same item characteristics, i.e. items that have high discrimination, low guessing and difficulty near the cut-off score give the most powerful statistical test. An efficiency study shows how many times more items are needed if nonoptimal items are used instead of optimal items in order to get the same power in the test.

  • 272.
    Wiberg, Marie
    Umeå University, Faculty of Social Sciences, Educational Measurement. Statistics.
    Changes in the Swedish driving-license test?: Using the GDE framework2007In: Proceedings from the conference: The GDE-model as a guide in driver training and testing: Umeå, May 7-8, 2007, 2007Conference paper (Other academic)
  • 273.
    Wiberg, Marie
    Umeå University, Faculty of Social Sciences, Statistics. Educational Measurement.
    Classical test theory vs. item response theory: An evaluation of the theory test in the Swedish driving-license test2004Report (Other academic)
    Abstract [en]

    The Swedish driving-license test consists of a theory test and a practical road test. The aim of this paper is to evaluate which Item Response Theory (IRT) model among the one (1PL), two (2PL) and three (3PL) parameter logistic IRT models that is the most suitable to use when evaluating the theory test in the Swedish driving-license test. Further, to compare the chosen IRT model with the indices in Classical Test Theory (CTT). The theory test has 65 multiple-choice items and is criterionreferenced. The evaluation of the models were made by verifying the assumptions that IRT models rely on, examining the expected model features and evaluating how well the models predict actual test results. The overall conclusion from this evaluation is that 3PL model is preferable to use when evaluating the theory test. By comparing the indices from CTT and IRT it was concluded that both give valuable information and should be included in an analysis of the theory test in the Swedish driving-license test.

  • 274.
    Wiberg, Marie
    Umeå University, Faculty of Social Sciences, Department of Educational Measurement. Umeå University, Faculty of Social Sciences, Department of Statistics.
    Datoriseringen av teoriprovet: En beskrivning av effekter utifrån ett antal statistiska indikatorer1999Report (Other academic)
  • 275.
    Wiberg, Marie
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Differential Item Functioning in Mastery tests: A comparison of three methods using real data2009In: International Journal of Testing, ISSN 1530-5058, E-ISSN 1532-7574, Vol. 9, no 1, p. 41-59Article in journal (Refereed)
    Abstract [en]

    The aim of this study was to examine log linear modelling (LLM) compared with logistic regression (LR) and Mantel-Haenszel (MH) test for detecting Differential Item Functioning (DIF) in a mastery test. The three methods were chosen because they have similar components. The results showed fairly high matching percentages together with high correlations among the methods regarding size of DIF. The MH approach yielded more conservative results than both LR and LLM. LLM and LR were fairly consistent with each other. The LLM has the advantage of dividing the test scores into certain intervals, which is of special interest in mastery tests. This partition of test scores was also tried with LR and MH with different results.

  • 276.
    Wiberg, Marie
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Estimation of try-out item parameters in computerized achievement testsManuscript (Other (popular science, discussion, etc.))
  • 277.
    Wiberg, Marie
    Umeå University, Faculty of Social Sciences, Educational Measurement. Statistics.
    Gender differences in the Swedish driving-license test2006In: Journal of Safety Research, ISSN 0022-4375, Vol. 37, no 3, p. 285-291Article in journal (Refereed)
    Abstract [en]

    Male drivers between 18–24 years are involved in many more motor-vehicle accidents than female drivers in Sweden. Is there

    also a gender difference in test performance in the Swedish driving-license tests? Is it a real difference or do any of the items display

    differential item functioning (DIF) with respect to gender or educational background? Method: Two random samples of test-takers'

    performance on the Swedish theory driving-license test were examined with descriptive and inference statistics. Results: Female test-takers

    performed better than male test-takers on the theory test, which is a real difference since no items displayed DIF with respect to gender. There

    was no gender difference in the practical test. The gender difference in the theory test could not be fully explained by differences in

    educational background. Conclusions: Although everyone who passes the driving-license test has fulfilled the goals stated in the curriculum,

    it is possible that the gender differences in the theory test are consistent over time, and therefore might be one factor that explains why male

    drivers are over-represented in traffic accidents. Impact on Industry: Gender differences appear persistent over time. Therefore, by improving

    male test takers' result on the test may lead to positive effects on the male test takers future driving career.

  • 278.
    Wiberg, Marie
    Umeå University, Faculty of Social Sciences, Department of Statistics. Umeå University, Faculty of Social Sciences, Department of Educational Measurement.
    Klassisk och modern testteori: analys av det teoretiska och praktiska körkortsprovet2004Report (Other academic)
    Abstract [en]

    The driving-license test in Sweden consists of a theory test and a practical test. The purpose of these two tests is to examine if the learner driver has the knowledge and abilities stated in the curriculum that the Swedish National Road Administration has issued. The purpose with this study is to examine the performance of the test takers, on both the theory test and the practical test over time and depending on the drivers´ education, age, gender and repeated test taking. Further to examine the links between the two tests and to examine the theory test with both classical test

    theory and item response theory.

    The results show that more test takers fail both the theory and the practical test than previous studies have shown and the test takers have more faults in their competence. The test takers’ educational background, i.e. if they have attended driving school or not, have a strong effect on the test scores in both tests. Test takers’ that have attended driving school perform better on both the practical and the theory test. There is a connection between the two tests such that test takers with high score on the theory test perform better on the practical test. There exists gender difference in the theory test such that women perform better than the men. An age difference is present though younger test-takers perform better than older test-takers on both tests. The number of test-takers that repeat the tests since they fail has increased over the years. The analysis with item response theory shows another side of the result than classical test theory which make it appropriate to use when analysing tests. For example, the theory test mainly examines test takers with low abilities and there are a number of problematic items in the test.

  • 279.
    Wiberg, Marie
    Umeå University, Faculty of Social Sciences, Educational Measurement. Statistics.
    Measuring and detecting differential item functioning in criterion-referenced licensing test: A theoretic comparison of methods2007Report (Other academic)
    Abstract [en]

    The validity of a measurement instrument depends on the quality of the items included in the instrument. The overall aim was to compare methods for detecting and measuring differential item functioning, DIF, in order to find a suitable method for examining DIF in a dichotomously scored criterion-referenced licensing test. The methods were discussed with respect to whether they are parametric, the nature of the matching score, if they can handle dichotomously and polytomously scored items, if they can test and/or measure DIF, and if they can detect both uniform and non-uniform DIF. The methods were also discussed with respect to whether they could handle the cut-off score in particular and the sample size requirements. The results show that there is not one method that can be recommended because many of them rely on strong assumptions which need to be examined and fulfilled before they can be recommended. It was recommended that an empirical study comparing the Mantel-Haenszel, logistic regression, log linear models and an IRT method is performed. Finally, the concluding remarks provide a discussion of guidelines for what to do if an item displays DIF in a test.

  • 280.
    Wiberg, Marie
    Umeå University, Faculty of Social Sciences, Statistics.
    Målrelaterade och normrelaterade prov: En teoretisk granskning av vilka statistiska tekniker som kan användas för att beskriva uppgifternas kvalitet och provets reliabilitet1999Report (Other academic)
  • 281.
    Wiberg, Marie
    Umeå University, Faculty of Social Sciences, Educational Measurement. Statistics.
    Ny kursplan och nya former för examination: Jämförelse av förarprovet för körkort B före och efter förändringen2007Report (Other academic)
    Abstract [sv]

    A new curriculum for the driving education system in Sweden was introduced on March 1, 2006. The purpose of this study was to evaluate a new driving-license test model which was tried during six months on three locations in Sweden. The test takers performance on the drivinglicense test after the changes was compared with their performance before the changes in the practical and the theory test. Further, the old and new practical tests and the old and new theory tests were compared.

    The result showed that the test takers improved their results on the driving-license test both on the theory and the practical test compared with the previous time period. Further, the integration between the theory test and the practical test was more evident than before. Test takers who had attended traffic school in general performed better both on the theory test and the practical test than test takers who had not attended driving school.

    The new theory test worked well, but the new practical test was somewhat problematic. One problem was that if a test taker had more than one competence shortage on the practical test, these could not be related to which traffic situations they had occurred in using the new form, i.e. the new practical test. Therefore, the new form needs to be revised before it should be used in large-scale assessment.

  • 282.
    Wiberg, Marie
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Om konsten att mäta kunskap med fokus på ekvivalering2009In: Thule : norrländsk kulturtidskrift, ISSN 0280-8692, p. 73-84Article in journal (Other academic)
  • 283.
    Wiberg, Marie
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Optimal sequential computerized knowledge tests2003In: Proceedings EYSM'03: September 21-26, 2003 Ovronnaz, Switzerland, 2003Conference paper (Refereed)
  • 284.
    Wiberg, Marie
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Optimal sequential mastery testsManuscript (preprint) (Other academic)
  • 285.
    Wiberg, Marie
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Sequential computerized mastery tests: three simulation studies2006In: International Journal of Testing, ISSN 1530-5058, E-ISSN 1532-7574, Vol. 6, no 1, p. 41-55Article in journal (Refereed)
    Abstract [en]

    A simulation study of a sequential computerized mastery test is carried out with items modeled with the 3 parameter logistic item response theory model. The examinees' responses are either identically distributed, not identically distributed, or not identically distributed together with estimation errors in the item characteristics. The simulations indicated that the observed results from the operating characteristic function differ significantly from the theoretical results, which is probably due to the use of an approximation formula. The mean number of items in a test, the distribution of test length, and the variance depend highly on how well we know the true values of the item characteristics and whether they are identically distributed or not.

  • 286.
    Wiberg, Marie
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Teaching statistics in integration with psychology2009In: Journal of Statistics Education, ISSN 1069-1898, E-ISSN 1069-1898, Vol. 17, no 1Article in journal (Refereed)
    Abstract [en]

    The aim was to revise a statistics course in order to get the students motivated to learn statistics and to integrate statistics more throughout a psychology course. Further, we wish to make students become more interested in statistics and to help them see the importance of using statistics in psychology research. To achieve this goal, several changes were made in the course. The theoretical framework to motivate teaching method changes was taken from the statistics education literature together with the ideas of student-centered learning and Kolb’s learning circle. One of the changes was to give the students research problems in the beginning of the course that were used throughout the course and which they should be able to solve at the end of the course. Other changes were to create a course webpage and to use more computer-based assignments instead of assignments with calculators. The students’ test results and their answers on the Survey of Attitudes Toward Statistics, SATS, (Schau, Stevens, Dauphinee, & Del Vecchio, 1995) together with course evaluations showed that by changing the course structure and the teaching, students performed better, and were more positive towards statistics even though statistics was not their major.

  • 287.
    Wiberg, Marie
    Umeå University, Faculty of Social Sciences, Department of Educational Measurement. Umeå University, Faculty of Social Sciences, Department of Statistics.
    Uppgiftsbank för körkortsprovets teoretiska prov: Relationen mellan utformningen, exponeringen och provtypen2002Report (Other academic)
  • 288.
    Wiberg, Marie H.
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Computerized achievement tests: sequential and fixed length tests2003Doctoral thesis, comprehensive summary (Other academic)
    Abstract [en]

    The aim of this dissertation is to describe how a computerized achivement test can be constructed and used in practice. Throughout this dissertation the focus is on classifying the examinees into masters and non-masters depending on their ability. However, there has been no attempt to estimate their ability.

    In paper I, a criterion-referenced computerized test with a fixed number of items is expressed as a statistical inference problem. The theory of optimal design is used to find the test that has the strongest power. A formal proof is provided showing that all items should have the same item characteristics, viz. high discrimination, low guessing and difficulty near the cutoff score, in order to give us the most powerful statistical test. An efficiency study shows how many times more non-optimal items are needed if we do not use optimal items in order to achieve the same power in the test.

    In paper II, a computerized mastery sequential test is examined using sequential analysis. The focus is on examining the sequential probability ratio test and to minimize the number of items in a test, i.e. to minimize the average sample number function, abbreviated as the ASN function. Conditions under which the ASN function decreases are examined. Further, it is shown that the optimal values are the same for item discrimination and item guessing, but differ for item difficulty compared with tests with fixed number of items.

    Paper III presents three simulation studies of sequential computerized mastery tests. Three cases are considered, viz. the examinees' responses are either identically distributed, not identically distributed, or not identically distributed together with estimation errors in the item characteristics. The simulations indicate that the observed results from the operating characteristic function differ significantly from the theoretical results. The mean number of items in a test, the distribution of test length and the variance depend on whether the true values of the item characteristics are known and whether they are iid or not.

    In paper IV computerized tests with both pretested items with known item parameters, and try-out items with unknown item parameters are considered. The aim is to study how the item parameters for try-out items can be estimated in a computerized test. Although the unknown examinees' abilities may act as nuisance parameters, the asymptotic variance of the item parameter estimators can be calculated. Examples show that a more reliable variance estimator yields much larger estimates of the variance than commonly used variance estimators.

  • 289.
    Wiberg, Marie
    et al.
    Umeå University, Faculty of Social Sciences, Educational Measurement. Statistics.
    Henriksson, Widar
    Umeå University, Faculty of Social Sciences, Educational Measurement.
    Metoder för kravgränssättning: En teoretisk granskning samt diskussion av lämplig metod för ett målrelaterat certifieringsprov av typ körkortsprovets teoriprov2000Report (Other academic)
    Abstract [sv]

    The main purpose of this study was to consider standard setting in a theoretical perspective. The aim was to describe the method categories which are used to set a standard, and to examine which category is the most suitable for a criterion-referenced licensure test. Criterion-referenced tests are used to ascertain an examinee’s status with respect to a well-defined criterion. A licensure test is administrated at the end of an education and covers the whole course. Another purpose of the study was to examine which method, in the most suitable category of methods, should be used to set a standard on a criterion-referenced licensure test like the Swedish driving licence test. The study describes how the standards are set in the old and new Swedish theoretical driving licence tests. The criteria used to decide the most suitable method were technical adequacy and practicability, each defined in several ways, and some specific criteria since it is a licensure test. The theoretical examination showed that it is not possible to find one single method that is best in all possible situations. The conclusion was that either Informed Judgement or Iterative Angoff was the most preferable method for setting a standard on the Swedish driving licence test.

  • 290.
    Wiberg, Marie
    et al.
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Rolfsman, Ewa
    Umeå University, Faculty of Social Sciences, Department of applied educational science.
    School-effectiveness in mathematics in Sweden compared with countries in Europe and Asia-Pacific2010Conference paper (Refereed)
    Abstract [en]

    There is an increased focus on educational quality and learning outcomes in Sweden, especially since international comparative studies like TIMSS has put the limelight on declining study results in core subject like mathematics. School-effectiveness is therefore an urgent issue. The aim of this study is to identify factors that contribute to the explanation why some schools is effective but others are less effective in terms of the students’ academic achievement in mathematics on TIMSS. This is conducted by examining Sweden in contrast with school-effectiveness in some high achieving countries in Europe and Asia-Pacific. The framework of multilevel analysis was used, since it is important to separate the effect of school-level variables from the effect of home environment and to take care of the sampling design used in TIMSS 2007. The results show that different educational system identifies different school level factors when controlling for home background. Noteworthy is that very few school level factors were significant in the full models. The single most important factor for school- effectiveness appears to be student behavior in school. Offering enrichment or remedial in mathematic seem to have different effects in different countries. Professional development opportunities for the teachers and the use of different incentives do not seem to have any effect in any of the countries, except in Chinese-Taipei. Disappointingly, we were unable to detect any significant school level factors in Sweden. A possible reason for this is that we excluded variables that have been altered in any of the countries, since we do believe that such changes might be a threat to the validity. Hence, it is possible that important questions in the Swedish context were excluded.

  • 291.
    Wiberg, Marie
    et al.
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Sundström, Anna
    Umeå University, Faculty of Social Sciences, Department of Educational Measurement.
    Implications of range restriction: a comparison of two approaches for correcting correlation in range restriction.2009In: Practical Assessment, Research & Evaluation, ISSN 1531-7714, E-ISSN 1531-7714, Vol. 14, no 5, p. 1-9-Article in journal (Refereed)
    Abstract [en]

    A common problem in predictive validity studies in the educational and psychological fields, e.g. in educational and employment selection, is restriction in range of the predictor variables. There are several methods for correcting correlations for restriction of range. The aim of this paper was to examine the usefulness of two approaches to correcting for range restriction; Thorndike’s case 2 correction and ML estimates obtained from the EM algorithm, by comparing the corrected correlations with the correlation from an unrestricted sample. The unrestricted sample consisted of examinees who took the practical Swedish driving-license test regardless of their result on the theory test. Examinees that passed the theory test and took the practical test were regarded as a restricted sample. The result provided empirical support for the appropriateness of Thorndike’s case 2 correction method. Although using the EM algorithm yielded a good estimate of the correlation in the unrestricted sample, further studies are needed on this topic.

  • 292.
    Wiberg, Marie
    et al.
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    van der Linden, Wim W.J.
    Local linear observed-score equating2011In: Journal of educational measurement, ISSN 0022-0655, E-ISSN 1745-3984, Vol. 48, no 3, p. 229-254Article in journal (Refereed)
    Abstract [en]

    Two methods of local linear observed-score equating for use with anchor-test and single-group designs are introduced. In an empirical study, the two methods were compared with the current traditional linear methods for observed-score equating. As a criterion, the bias in the equated scores relative to true equating based on Lord’s (1980) definition of equity was used. The local method for the anchor-test design yielded minimum bias, even for considerable variation of the relative difficulties of the two test forms and the length of the anchor test. Among the traditional methods, the method of chain equating performed best. The local method for singlegroup designs yielded equated scores with bias comparable to the traditional methods. This method, however, appears to be of theoretical interest because it forces us to rethink the relationship between score equating and regression.

  • 293.
    Wiklund, Stig-Johan
    Umeå University, Faculty of Social Sciences, Statistics.
    Control Charts and Process Adjustments1994Doctoral thesis, monograph (Other academic)
  • 294.
    Wolming, Simon
    et al.
    Umeå University, Faculty of Social Sciences, Educational Measurement.
    Wiberg, Marie
    Umeå University, Faculty of Social Sciences, Educational Measurement. Statistics.
    The Swedish driver licensure examination: Exploration of a two-stage model2004In: Journal of Safety Research, ISSN 0022-4375, Vol. 35, no 5, p. 491-495Article in journal (Refereed)
    Abstract [en]

    Introduction: Sweden, together with several countries in the world, has no compulsory educational requirement prior to the drivers license examination. Having a valid license examination (i.e., a knowledge test and a driving test) is therefore important in order to ensure competent drivers. Method: The aim of this paper is to explore the possibility of a two stage testing approach for the Swedish licensure examination. In order to explore the relationship between the knowledge test and the road test, a descriptive and a logistic regression approach were carried out. Results: The result indicates that how the driving candidates perform on the road test depends on their performance on the knowledge test. Therefore, the Swedish drivers licensing examination is suitable for a two-stage testing model.

  • 295.
    Wretman, Jan-Håkan
    Umeå University, Faculty of Social Sciences, Statistics.
    Om inferens vid ändliga populationer under superpopulationsantagande1970Licentiate thesis, monograph (Other academic)
  • 296.
    Wållberg-Jonsson, Solveig
    et al.
    Umeå University, Faculty of Medicine, Department of Public Health and Clinical Medicine, Reumatology.
    Johansson, Helene
    Öhman, Marie-Louise
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Rantapää-Dahlqvist, Solbritt
    Umeå University, Faculty of Medicine, Department of Public Health and Clinical Medicine, Reumatology.
    Extent of inflammation predicts cardiovascular disease and overall mortality in seropositive rheumatoid arthritis: A retrospective cohort study from disease onset1999In: Journal of Rheumatology, ISSN 0315-162X, E-ISSN 1499-2752, Vol. 26, no 12, p. 2562-2571Article in journal (Refereed)
    Abstract [en]

    OBJECTIVE: To identify predictors for cardiovascular disease (CVD) and for overall survival in patients with rheumatoid arthritis (RA) followed from disease onset.

    METHODS: A retrospective cohort of patients with seropositive RA and disease onset between 1974 and 1978 (n = 211) was followed up at the end of 1995. Potential predictors for CVD, as measured by "the first cardiovascular event," and for overall survival were registered. The predictors were identified by extended Cox regression models.

    RESULTS: In simple Cox regression analysis, male sex, higher age at disease onset, HLA-B27, high disease activity, corticosteroid treatment early in disease, and hypertension significantly increased risk of cardiovascular event. Higher educational level, extensive disease modifying antirheumatic drug (DMARD) treatment, and corticosteroids > or =1 yr before event decreased the risk. In multiple Cox regression analysis, male sex, high age at disease onset, hypertension, higher haptoglobin level at disease onset, and corticosteroid treatment early in disease increased risk of CVD. In a multiple model comprising only patients with CVD, corticosteroids delayed the event. A high last registered erythrocyte sedimentation rate (ESR) value before event increased CVD risk, in particular when early in disease progression. Decreased life span was predicted by higher age at disease onset, male sex, low education level, high disease activity, hypertension, and CVD. HLA-B27 was associated with decreased life span, as was early, but not extensive corticosteroid treatment. DMARD treatment was associated with decreased mortality risk, as was the presence of joint prosthesis. In multiple regression, male sex, higher age at disease onset, atlantoaxial subluxation early in disease, hypertension, and cardiovascular event increased mortality. A high last registered ESR value before event or death added to that risk.

    CONCLUSION: The study emphasizes the importance of inflammation as an important risk indicator for CVD and mortality in RA. The positive impact of disease activity reducing treatment on CVD risk and survival is suggested.

  • 297.
    Wållberg-Jonsson, Solveig
    et al.
    Umeå University, Faculty of Medicine, Department of Public Health and Clinical Medicine, Reumatology.
    Öhman, Marie-Louise
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Rantapää-Dahlqvist, Solbritt
    Umeå University, Faculty of Medicine, Department of Public Health and Clinical Medicine, Reumatology.
    Cardiovascular morbidity and mortality in patients with seropositive rheumatoid arthritis in Northern Sweden.1997In: Journal of Rheumatology, ISSN 0315-162X, E-ISSN 1499-2752, Vol. 24, no 3, p. 445-451Article in journal (Refereed)
    Abstract [en]

    OBJECTIVE: To investigate the overall and the cardiovascular mortality in rheumatoid arthritis (RA) in Northern Sweden. To analyze the effect of traditional risk factors and factors associated with rheumatoid disease and its treatment on the progression of cardiovascular disease (CVD) and on mortality by all causes. METHODS: A cohort of 606 patients with seropositive RA were followed from 1979 to the end of 1994 or to the death of the patient. Standardized mortality ratio and survival curves were estimated with the population of Vasterbotten as reference. Sex, age at disease onset, treatment with corticosteroids, use of disease modifying antirheumatic drugs (DMARD) and hormone replacement therapy (HRT), hypertension, diabetes mellitus, HLA types, and cause of death were recorded from disease onset. Cox's proportional hazards regression was used to identify important predictors for death and cardiovascular event during followup. RESULTS: The standardized mortality ratio in both sexes was significantly higher (1.57) for all underlying causes together, for CVD (1.46) and for ischemic heart disease (IHD) (1.54) compared to the reference population. The death rate increased over time. In multiple Cox regression analyses, male sex, higher age at disease onset, and former cardiovascular event increased the death rate. Male sex, high age at disease onset, and hypertension increased the risk of cardiovascular event. Diabetes mellitus, treatment with corticosteroids, DMARD, or HRT did not influence the risks of death or first cardiovascular event. CONCLUSION: The overall mortality and death due to CVD and IHD were in both sexes increased in seropositive RA. Male sex and high age at disease onset predicted death and cardiovascular event. Except for hypertension, which increased the risk for cardiovascular event, neither of these traditional cardiovascular risk factors nor corticosteroid treatment influenced mortality by all causes or by cardiovascular event.

  • 298.
    Wållberg-Jonsson, Solveig
    et al.
    Umeå University, Faculty of Medicine, Public Health and Clinical Medicine.
    Öhman, Marie-Louise
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Rantapää-Dahlqvist, Solbritt
    Umeå University, Faculty of Medicine, Public Health and Clinical Medicine.
    Which factors are related to the presence of atherosclerosis in rheumatoid arthritis?2004In: Scandinavian Journal of Rheumatology, ISSN 0300-9742, E-ISSN 1502-7732, Vol. 33, no 6, p. 373-9Article in journal (Refereed)
  • 299.
    Öhman, Marie-Louise
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    A Monte Carlo study of some censored data Wilcoxon rank tests1990In: Biometrical Journal, ISSN 0323-3847, E-ISSN 1521-4036, Vol. 32, no 6, p. 721-735Article in journal (Refereed)
    Abstract [en]

    The variance estimators usually applied for the generalized censored data Wilcoxon rank tests by Gehan and Peto & Prentice, are heavily biased in unbalanced problems. This paper reports the results of a Monte Carlo simulation study, where jackknifing is used to construct estimators of variance. Size, power and variance properties are compared for five variance estimators, when using different combinations of group sizes, failure and censoring patterns. The variance estimators are the permutational, the conditional permutational and the jackknife variance estimators for the statistic of Gehan and the asymptotic and the jackknife variance estimators for the statistic of Peto & Prentice. It appears that observed size, power and variance properties may be improved by using the jackknife variance estimator, when comparing to the variance estimators usually applied.

  • 300.
    Öhman, Marie-Louise
    Umeå University, Faculty of Social Sciences, Department of Statistics.
    Aspects of analysis of small-sample right censored data using generalized Wilcoxon rank tests1994Doctoral thesis, monograph (Other academic)
    Abstract [en]

    The estimated bias and variance of commonly applied and jackknife variance estimators and observed significance level and power of standardised generalized Wilcoxon linear rank sum test statistics and tests, respectively, of Gehan and Prentice are compared in a Monte Carlo simulation study. The variance estimators are the permutational-, the conditional permutational- and the jackknife variance estimators of the test statistic of Gehan, and the asymptotic- and the jackknife variance estimators of the test statistic of Prentice. In unbalanced small sample size problems with right censoring, the commonly applied variance estimators for the generalized Wilcoxon rank test statistics of Gehan and Prentice may be biased. In the simulation study it appears that variance properties and observed level and power may be improved by using the jackknife variance estimator.

    To establish the sensitivity to gross errors and misclassifications for standardised generalized Wilcoxon linear rank sum statistics in small samples with right censoring, the sensitivity curves of Tukey are used. For a certain combined sample, which might contain gross errors, a relatively simple method is needed to establish the applicability of the inference drawn from the selected rank test. One way is to use the change of decision point, which in this thesis is defined as the smallest proportion of altered positions resulting in an opposite decision.

    When little is known about the shape of a distribution function, non-parametric estimates for the location parameter are found by making use of censored one-sample- and two-sample rank statistics. Methods for constructing censored small sample confidence intervals and asymptotic confidence intervals for a location parameter are also considered. Generalisations of the solutions from uncensored one-sample and two-sample rank tests are utilised. A Monte-Carlo simulation study indicates that rank estimators may have smaller absolute estimated bias and smaller estimated mean squared error than a location estimator derived from the Product-Limit estimator of the survival distribution function.

    The ideas described and discussed are illustrated with data from a clinical trial of Head and Neck cancer.

34567 251 - 300 of 303
CiteExportLink to result list
Permanent link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf