Statistical Methods in Ecology with Application to Tree Growth
Description: Since 1990 students working with Dr. Shea have studied the growth and survival patterns of deciduous and conifer trees planted as part of the St. Olaf Natural Habitat Restoration Program. Little information is available on early growth patterns of trees. In 2010 research students will be collecting data on tree size and mortality of trees studied over a 20 year time span in a maple-basswood restored forest. Maple-basswood forests are the dominant forest type in southeastern Minnesota and the type of forest we want to have in most of out forest restoration areas. Longitudinal data will make it possible for growth curves to be established for six to eight different species. Statistical models could be developed to compare actual and predicted growth patterns. Statistics and ecology students will work together to better understand variation in tree growth patterns.
These data will be useful in evaluating the success of the restoration project and in making recommendations for future restoration projects. Measurements of tree height and diameter allow us to compare growth patterns of different species and evaluate conditions under which each species grows better, thus contributing to a better understanding of the process of early forest succession.
Domain Expert: Kathy Shea, Biology
Curriculum-Based Measurement in Writing: Longitudinal Research with Elementary School Learners
Description: For the past several years, we have collected longitudinal writing data (curriculum-based measurement: CBM) with 5th grade English language learners and native English speakers in St. Paul, MN. We are hoping to use the CBM data to determine whether or not the students make adequate writing progress and whether changes in instruction result in increased slopes in CBM data. Possible statistics needs: the use of hierarchical linear models to analyze longitudinal data as well as creating “tables of probable success” to predict which students might pass their MCA spring tests based on fall CBM data. Additionally, we would like to compare this year’s results with the results obtained in previous studies that looked solely at the writing of English language learners.
Domain Expert: Heather Campbell, Education
Statistical Applications in Social Science Research
Description: Each year, students in SOAN 371 (Foundations of Social Science Research) conduct a random-sample survey of several hundred St. Olaf students on a topic of general interest, analyze the data, and present their results in various forums. CIR Fellows would serve as consultants to research groups from SOAN 371, assisting in data analysis and learning the consultant role. Recent years’ topics have included student satisfaction, use of electronic communication, and close relationships. This coming year, we will again analyze survey data from St. Olaf College students. Topic options include health and substance use, social/political values and their impact on students’ lives, and social networks. Past CIR Fellows have joined us to present research at a professional conference and would be invited to do so again.
Domain Expert: Ryan Sheppard, Sociology
The effectiveness of influence strategies in romantic relationships
Description: Previous research has provided evidence that of three global influence strategies that partners use to change their partner’s behaviors (making reference to desired behaviors that strengthen the relationship, promising rewards or punishments, and using logical or rational appeals), the only strategy that has been associated with observable opinion change entails appealing to relationship-based concerns. This correlational evidence, however, does not allow us to determine whether this strategy was causally related to opinion change, both public and private. Upon completion of an experimental study, we will use the data to determine whether any of the three global influence strategies is effective in changing privately held attitudes.
Domain Expert: Minda Orina, Psychology
Will You Love Me When I’m Gone? Marketing Planned Giving to St. Olaf Alumni
Description: St. Olaf College’s Advancement and College Relations division is currently focusing on increasing membership to our planned giving society, the Manitou Heights Society. We want to measure the effectiveness of various marketing strategies to several target populations to determine the most effective and efficient means of encouraging our constituents to support the college through various planned giving vehicles. We would also like to use analytics to evaluate our current Manitou Heights Society members and begin to build predictive models of likely planned giving candidates.
Domain Expert: Jackie Henry, Dir. Advancement Services
Fact-Checking Principles of Economics
Description: Many theories and “laws” are put forth in introductory economics classes as though they are invariant. However, the most popular texts frequently offer no empirical evidence at all in support of their theories nor do they attempt to deal with controversies in economics by appealing to the data (which is Latin for “facts”). From the multiplier model to the determination of interest rates to the conduct of monetary policy, the introductory texts are ripe for fact checking. I see the project beginning with a review of the most popular texts with an eye towards finding those topics most amenable to simple data-based testing that can be conveyed to introductory students. Next, we’ll look at the literature and techniques needed for analysis. Then, we’ll “crunch the numbers” and see what they say. Most of the focus will be on macroeconomics and time-series analysis. I expect that our work can be submitted to the Journal of Economic Education and presented at conferences (funding permitting) such as Minnesota Economic Assoc. (April 2011) and Western Econ. Assoc. (July 2011, San Diego).
Domain Expert: Tony Becker, Economics
Global Health in Geneva
Description:
One of the projects we will be working on in Geneva involves the estimation of foodborne disease. Foodborne diseases encompass a wide spectrum of illnesses and are a growing public health problem world-wide. They are the result of ingestion of contaminated food stuffs, and range from diseases caused by a multitude of microorganisms to those caused by chemical hazards. The contamination of food may occur at any stage in the process from food production to preparation (‘farm to fork’) and result from environmental contamination, including pollution of water, soil and/or air. It is for this reason that this burden of disease work is coordinated with other WHO activities in this area, including those on chemicals, water, sanitation and hygiene. (http://www.who.int/foodborne_disease/burden/en/)
The most common clinical presentation of foodborne diseases takes the form of gastrointestinal symptoms but such diseases can also include neurological, gynaecological, immunological and other symptoms. Multi-organ failure and even cancers may result from the ingestion of contaminated food stuffs, thus carrying a considerable disability as well as mortality burden. Surveillance data from developed countries and sentinel sites indicate a high disease burden for foodborne diseases caused by microorganisms alone. Surveillance data, however, tend to show only the tip of the clinical iceberg and cannot sufficiently describe true disease burden. For affected persons to feature in such health statistics, they not only have to seek medical care, provide a specimen for laboratory investigation, and test positive on laboratory methods but also be reported to the relevant health authorities.
To circumvent the problems posed by such under-reporting and describe disease burden more adequately, a number of innovative and creative approaches have been used in recent years for some foodborne diseases from various causes. These include the use of active surveillance and field studies, risk assessment methods, and epidemiological disease modelling. For other foodborne diseases, however, including some zoonoses and diseases cause by chemical hazards, no such data or studies exist.
In order to estimate disease burden adequately and provide comprehensive information for policy makers it is important to move beyond the mere quantification of morbidity and mortality and describe burden in a summary measure that includes elements of severity and duration of disease, as well as resulting disability. One such summary indicator is the Disability Adjusted Life Year (DALY) which has been widely used by WHO and others in the field of burden of disease estimation.
Domain Expert: Dr. Claudia Stein, Epidemiologist at the World Health Organization