GATHERING GRAND SLAM STATISTICS ON TENNIS PLAYERS: The Grand Slam tennis program is a statistical model that allows you to simulate matches between former winners of grand slam tournaments (Australian Open, French Open, Wimbledon, and the U.S. Open). The design and programming of the statistical model relies on statistics that are researched and gathered from all Grand Slam Tennis singles events occurring from 1920 to the present day. You can download a free shareware copy of the "Grand Slam Tennis" program from this web site.
ARTS PARTICIPATION ESTIMATES: One of the first survey research projects that I worked on was the 1982 Survey of Public Participation in the Arts (SPPA). This was the first large national study the produced estimates of how often people participated in the arts. Since then, I have been involved in helping design many of the later SPPA studies (1982, 1985, 1992, 1997, 2002, 2008,2012). Here is the link to an article on the accuracy of spouse/partner proxy reporting that was part of the 2008 and 2012 survey design.
WINE RATINGS: Wine ratings (100-point scale) for this web site began in 1994. A sampling of at least one full glass of wine is needed before a wine receives a rating. You can see how different grapes, vineyards, and regions rate by checking out the wine ratings on this web site.
CONDUCTING EFFECTIVE SURVEYS WITH IMMIGRANT AND REFUGEE POPULATIONS: This research examines the data collection challenges and complexities of conducting the Annual Survey of Refugees (ASR). The ASR is a telephone survey that collects information on 1,500 refugee families during their first five years after arrival in the U.S. It is the only scientifically collected source of national data on refugee's progress toward self-sufficiency and integration. It collects information in twenty languages, capturing a sizeable portion of the diversity in the U.S. refugee population. This research investigates some of the lessons learned on the ASR and how these lessons can be used to improve the methodology used on other nationally representative telephone surveys or cross-cultural surveys.
RESEARCH ON RESPONDENT'S ABILITY TO RECALL PAST EVENTS: Recently I have been conducting research on who has trouble reporting prior events. Here is a link to an article that looks at recall problems using data from the American Driving Study where people are asked to report the length of driving trips that they made yesterday.
RESEARCH ON WHETHER LOWER PROGRAM DOSAGE/TREATMENT INCREASES NONRESPONSE: It is clear from the literature (Annette Aigner, Ulrike Grittner, Heiko Becher, 2018) that higher nonresponse from control group respondents that get no treatment leads to the recruitment of a non-representative sample of controls in case-control studies. But, what about nonresponse from respondents receiving treatment who receive less than full treatment. If the response rate is lower for those receiving less than full treatment, then we could be getting a bias sample of respondents in the treatment group. Nonresponse usually entails using survey weights that adjust our sample to reduce potential bias and create a sample that is representative of the information available in the sampling frame. Although weighting adjustments could be helping, are there other weighting adjustments, sampling strategies, or survey design changes that could lead to a more representative sample when treatment varies.
Timothy Triplett's Home Page: