Do Won Kwak

Assessing the Impact of Blended Learning on Student Performance

This paper assesses quantitatively the impact on student performance of a blended learning experiment within a large undergraduate first-year course in statistics for business and economics students. We employ a difference-in-differences method, which controls for differences in student characteristics and course delivery method, to evaluate the impact of blended learning on student performance. Our results suggest that the impact of blended learning on student performance depends on whether the effect of blended learning is cumulative or not. Blended learning has no impact on student performance if learning is non-cumulative and only affects the performance on the quizzes associated with the material covered by blended learning. However, if learning is cumulative and impacts the performance for the whole course, then our results strongly suggest a strong, negative effect. Taken as a whole, these results provide a possible explanation for why most of the existing studies focusing on short online courses have obtained neutral or even positive results while nearly all the studies focusing on semester-length course tend to observe negative impacts from online learning.

The WTO puzzle, multilateral resistance terms and multicollinearity

Since Rose’s (2004) striking finding of negligible WTO trade effects, numerous studies have attempted to solve the so-called WTO puzzle. These studies have progressively improved model specifications to control for potential sources of bias, but they often lead to conflicting results. Multilateral resistance terms (MRTs) are considered to be one of the most crucial factors to be accounted for to avoid the omitted variable bias in the gravity model. What has gone unnoticed, however, is that the control for MRTs leads to near-prefect multicollinearity because of the structural relationship between the variables that measure the General Agreement on Tariffs and Trade/World Trade Organization (GATT/WTO) membership status of any country pairs. This multicollinearity contributes to the fragility of the WTO effect estimates, as it dramatically increases SEs. In this article, we explain how this multicollinearity arises and provide evidence of it.

It Is Much Bigger Than What We Thought: New Estimate of Trade Diversion

Conventional estimate of the trade diversion (TD) effect of preferential trade agreements (PTAs) tends to be much smaller than that of the trade creation (TC) effect. This paper ex- amines two sources of estimation bias of the TD effect. The first bias of TD arises from the difficulty in controlling for multilateral resistance and other unobserved time-varying country heterogeneity when estimating both TC and TD effects simultaneously due to perfect mul- ticollinearity. The second one is the underestimation of TD and arises from the failure to recognize that the concept of TD is inapplicable for a substantial proportion of PTAs and, thus, that the conventional method wrongly counts cases where trade cannot be diverted as cases where trade is not diverted. This paper corrects these two biases by using fixed effects and introducing a new measure of TD which provides a better mapping between the theoretical concept and data characteristics. Removing the two sources of bias leads to a twelve-fold in- crease in the estimate of the TD effect. It is found that the total TD effect is comparable to the TC effect in dollar terms.

New insights into an old problem – enhancing student learning outcomes in an introductory statistics course

ABSTRACT Many students enrolled in first year introductory statistics courses believe learning statistics is a waste of time and fear they will fail. In this study, we explored the impacts on learning outcomes for students in an introductory statistics course by allowing students to arbitrarily choose their own sequence of learning from three key learning activities, namely tutorials, Peer-Assisted Study Sessions and Computer-Managed Learning quizzes. Unlike the old regime where the learning activities followed a strict, rigid sequence, a new regime allowed students to freely choose when, where and how they engaged with the course learning activities. This allowed increased opportunities for students to receive relevant and timely feedback. Using a total of 1187 students enrolled in semester 2 of 2011, 2012 and 2013, data were collected on students’ scores from 7 assessment tasks. Our experimental design ensured as many course features as possible remained constant between the control cohorts (of 2011 and 2012) and the experimental cohort (2013), thereby avoiding potential sample selection problems. The findings showed student learning outcomes in the new regime improved significantly. Interestingly, the effects were found to be greatest in the lower percentile of the score distribution.

Leading by example: do the biggest and most generous donors influence others to give more health aid?

Purpose - – The purpose of this paper is to explore the interdependence between donor countries’ health aid expenditures. The specific form of interdependence considered is the leader effect, whereby an influential country has a positive leverage effect on other donor countries’ aid expenditure. The opposite case of a free-rider effect, whereby a single donor country has a negative leverage effect on its peers, is also considered. Design/methodology/approach - – Focusing on the identification of the leader effect avoids the estimation bias present in the identification of the peer group effect, due to endogenous social effect. The empirical analysis focuses on Development Assistance for Health provided by 20 OECD countries over the period of 1990-2009. Aid commitment and aid disbursement are distinguished. Findings - – When aid dynamics, country heterogeneity, and endogeneity are accounted for, there is no evidence that the biggest donor – the USA, or the most generous donors – Norway and Sweden, exhibit any leverage effects on other donor countries’ aid expenditures. Originality/value - – This is the first paper to examine the leader and free-rider effects in health aid provision as previous studies focus on peer effects. Any evidence of leader or free-rider effects (or the lack of it) adds to the understanding of international political economy especially in the area of foreign aid provision.