Statistical analysis of survey-based event history data with application to modeling of unemployment duration

Longitudinal surveys are increasingly used to collect event history data on person-specific processes such as transitions between labour market states. Surveybased event history data pose a number of challenges for statistical analysis. These challenges include survey errors due to sampling, non-response, attrition and measurement. This study deals with non-response, attrition and measurement errors in event history data and the bias caused by them in event history analysis. The study also discusses some choices faced by a researcher using longitudinal survey data for event history analysis and demonstrates their effects. These choices include, whether a design-based or a model-based approach is taken, which subset of data to use and, if a design-based approach is taken, which weights to use. The study takes advantage of the possibility to use combined longitudinal survey register data. The Finnish subset of European Community Household Panel (FI ECHP) survey for waves 1–5 were linked at person-level with longitudinal register data. Unemployment spells were used as study variables of interest. Lastly, a simulation study was conducted in order to assess the statistical properties of the Inverse Probability of Censoring Weighting (IPCW) method in a survey data context. The study shows how combined longitudinal survey register data can be used to analyse and compare the non-response and attrition processes, test the missingness mechanism type and estimate the size of bias due to non-response and attrition. In our empirical analysis, initial non-response turned out to be a more important source of bias than attrition. Reported unemployment spells were subject to seam effects, omissions, and, to a lesser extent, overreporting. The use of proxy interviews tended to cause spell omissions. An often-ignored phenomenon classification error in reported spell outcomes, was also found in the data. Neither the Missing At Random (MAR) assumption about non-response and attrition mechanisms, nor the classical assumptions about measurement errors, turned out to be valid. Both measurement errors in spell durations and spell outcomes were found to cause bias in estimates from event history models. Low measurement accuracy affected the estimates of baseline hazard most. The design-based estimates based on data from respondents to all waves of interest and weighted by the last wave weights displayed the largest bias. Using all the available data, including the spells by attriters until the time of attrition, helped to reduce attrition bias. Lastly, the simulation study showed that the IPCW correction to design weights reduces bias due to dependent censoring in design-based Kaplan-Meier and Cox proportional hazard model estimators. The study discusses implications of the results for survey organisations collecting event history data, researchers using surveys for event history analysis, and researchers who develop methods to correct for non-sampling biases in event history data.

From the seeds of rediscovered texts come new fields of inquiry

Kirjallisuusarvostelu

The duality of tactical thought : a study of how Swedish land forces' commanders view tactics in irregular warfare

This is a sociological study of the views of officers in the Swedish Army and its Amphibious Forces on tactics in Irregular Warfare (IW), in particular, Counterinsurgency (COIN). IW comprises struggles, where the military weaker part uses an indirect approach with smaller units and integrates the civilian and military dimensions in a violence spectrum including subversion, terrorism, Guerrilla Warfare and infantry actions. IW is the main armed warfare style in insurgencies. COIN is the combined political, military, economic, social and legal actions in counter insurgencies. Data has been collected by means of interviews with almost all (n =43) officers, who were either commanding battalions or rifle and manoeuvre companies while undergoing training for general warfare and international operations. The main theoretical and methodological inspiration is the traditional one for research on social fields, inaugurated by the French sociologist Pierre Bourdieu. The statistical technique used is Multiple Correspondence Analysis. As a background and context base, an inquiry inspired by the Begriffsgechichte (Conceptual History) tradition explores the genesis and development of understandings of the term Irregular Warfare. The research question is outlined as; “how can contemporary Swedish military thought on tactics in Irregular Warfare be characterized using descriptive patterns, mapped in relation to background factors and normative standards? The most significant findings are that there are two main opposing notions separating the officers’ views on tactics in Irregular Warfare: (1) a focus on larger, combat oriented and collectively operating military units versus smaller and larger, more intelligence oriented and dispersed operating units, and (2) a focus on military tasks and kinetic effects versus military and civilian tasks as well as “soft” effects. The distribution of these views can be presented as a two-dimensional space structured by the two axes. This space represents four categories of tactics, partly diverging from normative military standards for Counterinsurgency. This social space of standpoints shows different structural tendencies for background factors of social and cultural character, particularly dominant concerning military backgrounds, international mission experiences and civilian education. Compared to military standards for Counterinsurgency, the two tactical types characterized by a Regular Warfare mind-set stands out as counter-normative. Signs of creative thought on military practice and theory, as well as a still persistent Regular Warfare doxa are apparent. Power struggles might thus develop, effecting the transformation to a broadened warfare culture with an enhanced focus also on Irregular Warfare. The result does not support research results arguing for a convergence of military thought in the European transformation of Armed Forces. The main argument goes beyond tactics and suggests sociological analysis on reciprocal effects regarding strategy, operational art, tactics as well as leadership, concerning the mind-set and preferences for Regular, Irregular and Hybrid Warfare.

Statistical segmentation methods and color variance analysis of retinal images

In this research, the effectiveness of Naive Bayes and Gaussian Mixture Models classifiers on segmenting exudates in retinal images is studied and the results are evaluated with metrics commonly used in medical imaging. Also, a color variation analysis of retinal images is carried out to find how effectively can retinal images be segmented using only the color information of the pixels.

Utilization of Statistical Methods for Management in the Forest Industry

The strongest wish of the customer concerning chemical pulp features is consistent, uniform quality. Variation may be controlled and reduced by using statistical methods. However, studies addressing the application and benefits of statistical methods in forest product sector are scarce. Thus, the customer wish is the root cause of the motivation behind this dissertation. The research problem addressed by this dissertation is that companies in the chemical forest product sector require new knowledge for improving their utilization of statistical methods. To gain this new knowledge, the research problem is studied from five complementary viewpoints – challenges and success factors, organizational learning, problem solving, economic benefit, and statistical methods as management tools. The five research questions generated on the basis of these viewpoints are answered in four research papers, which are case studies based on empirical data collection. This research as a whole complements the literature dealing with the use of statistical methods in the forest products industry. Practical examples of the application of statistical process control, case-based reasoning, the cross-industry standard process for data mining, and performance measurement methods in the context of chemical forest products manufacturing are brought to the public knowledge of the scientific community. The benefit of the application of these methods is estimated or demonstrated. The purpose of this dissertation is to find pragmatic ideas for companies in the chemical forest product sector in order for them to improve their utilization of statistical methods. The main practical implications of this doctoral dissertation can be summarized in four points: 1. It is beneficial to reduce variation in chemical forest product manufacturing processes 2. Statistical tools can be used to reduce this variation 3. Problem-solving in chemical forest product manufacturing processes can be intensified through the use of statistical methods 4. There are certain success factors and challenges that need to be addressed when implementing statistical methods

Tietoperustaisten HRM-käytäntöjen, luottamuksen ja suorituskyvyn suhde suomalaisessa organisaatiokontekstissa

Henkilöstöjohtamisen käytännöillä on merkittävä vaikutus organisaation johtamiseen ja tätä kautta menestymiseen. Hyvin suunnitellut ja toteutetut käytännöt ovat osaltaan vaikuttamassa yrityksessä vallitsevaan luottamukseen. Hyvä sisäinen luottamustaso puolestaan heijastuu myös organisaation ulkoiseen luottamukseen. Tietoperustaiset HRM-käytännöt ovat keskeisessä roolissa juuri nyt, koska työelämä ja työn tekeminen ovat murrosvaiheessa ja muutosten nopeus kiihtyy entisestään tietointensiivisten työtehtävien lisääntyessä. HRM-käytäntöjä on tutkittu paljon aiempien vuosikymmenten aikana. Tietoperustaisten henkilöstöjohtamisen käytäntöjen tutkimusta on ollut jonkin verran 2000-luvun alun jälkeen. Henkilöstöjohtamisen käytäntöjen suhdetta luottamukseen ja suorituskykyyn on myös tutkittu. Sen sijaan vielä ei ollut tehty tutkimusta, jossa yhdistetään tietoperustaisten henkilöstöjohtamisen käytäntöjen, luottamuksen ja suorituskyvyn suhde. Tämän tutkimuksen tarkoitus on selvittää tietoperustaisten HRM-käytäntöjen, luottamuksen ja suorituskyvyn yhteyttä. Tutkimuksen pääpaino on tietoperustaisten henkilöstökäytäntöjen suhde luottamukseen. Tutkittavat tietoperustaiset henkilöstöjohtaminen käytännöt liittyvät rekrytointiin, osaamisen kehittämiseen, suoritusarviointiin ja palkitsemiseen. Tutkimukseen osallistui 246 Suomessa toimivaa yritystä. Kyselytutkimuksen lisäksi tutkimuksessa käytettiin taloudellisia tietokantoja yritysten suorituskykyä kuvaavien tunnuslukujen määrittämisessä. Tutkimus on määrällinen ja se perustuu tilastollisten menetelmien ja analyysien käyttöön. Tutkimuksessa saatiin selville, että tietoperustaisilla osaamisen kehittämisen ja suoritusarviointien HRM-käytännöillä on positiivinen vaikutus luottamukseen. Tietoon pohjautuvilla henkilöstöjohtamisen käytännöillä voidaan siis rakentaa parempaa luottamustasoa organisaatioissa.