Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: ????One of the paramount empirical issues regarding the mingling of policy and administration is the extent to which this intersection occurs because of the aggrandizing tendencies of bureaucracy. A second prominent concern of both theoretical and empirical interest is administrative accountability. What are the patterns of actual influence on the public administrator from what sources (or actors) and which sources/actors does the administrator prefer to exercise influence? ????This paper addresses these central questions by examining the views of state agency heads on the extent of actual and preferred influence exercised by various actors on bureaucratic initiatives. Using data from the 1984 and 1994 American State Administrators Project surveys, we assess state agency heads perceptions of the actual and preferred levels of influence exerted on major policy decisions. State agency heads prefer a higher level of influence by numerous institutional and other actors. Of the eight policy actors whose influence is probed, only for state legislators and state courts do administrators attribute too much influence. State administrators prefer greater influence for themselves and their agencies, among other policy actors. But greater self/agency influence is not associated with decreased influence of other actors. We find consistent positive-sum patterns across two decades. When factor analysis of actual and preferred influence is used to explore patterns of administrative accountability for the eight policy actors, several readily interpretable factors emerge. These dimensions of accountability reveal striking commonalities in agency head perceptions of institutional and external actor influences over two decades. The several accountability dimensions are remarkably consistent with the conceptual characterizations of public administrators' descriptive and normative obligations in a democratic society that have been articulated over the past six decades.

Publication Type: Conference Paper/Unpublished Manuscript
Abstract: Central to fiscal federalism is the idea that matching grants stimulate local spending more than lump-sum subsidies. The paper questions the conventional wisdom with empirical evidence from a change of grant regime in Danish local governments.

Publication Type: Short Research Paper
Abstract: I examined preservice teachers' (PST) coordination of representational units arising from the use of magnetic color cubes and tiles for representing various special numbers. The coordination and identification of these representational units of different - not necessarily hierarchical - types (multiplicative, additive, pseudo-multiplicative) appear to be important mathematical practices in dealing with problems involving representational quantities.

Publication Type: Conference Paper/Unpublished Manuscript
Review Method: Peer Reviewed
Abstract: In the calculus sequence the sum of an infinite series is usually defined as the limit of the sequence of partial sums, if it exists. But there are infinite series, such as Grandi’s series, which are not given a sum by this definition and for which there are good arguments in favor of a sum. In this talk, we will consider a discussion which the presenter uses in his calculus classes to explore the concept of a “better” definition based on historical approaches to the sum of an infinite series proposed by Cesàro and Euler.

Publication Type: Conference Paper/Unpublished Manuscript
Abstract: It is well known that a pair of 6-sided dice cannot be labeled so that the sums $\$2,\; 3,\; 4,\; \backslash ldots,\; 12\$$ are all equally likely. Edward J. Dudewicz and Ronald E. Dann showed that there is no way to label $\$m\$$ $\$n\$$-sided dice with integers from the set $\$\backslash \{1,\; 2,\; 3,\; \backslash ldots,\; n\backslash \}\$$ such that the resulting sums are all equally likely. We investigate a similar question, but we eliminate the condition that the dice only be labeled with integers from the set $\{1, 2, 3, \ldots, n\}$. Namely, when can $\$m\$$ $\$n\$$-sided dice be labeled with any nonnegative integers such that the $``standard\; sums"$ of $\$m,\; m+1,\; m+2,\; \backslash ldots,\; mn\$$ are all equally likely? Any labeling of dice which yields the standard sums with equal probability will be called an $\backslash emph\{equal\; labeling\}$. Even if we eliminate Dudewicz and Dann's condition that the dice labels come from the set $\$\backslash \{1,\; 2,\; 3,\; \backslash ldots,\; n\backslash \}\$$, pairs of dice will always fail to have an equal labeling, unless the dice are one-sided. However, there are infinitely many $\$m\$$ such that $\$m\$$ 6-sided dice afford an equal labeling. We characterize all such $\$m\$$ and discuss the
resulting dice labels.