June 2000
[updates past files:  January 1997, updated October 1997]

The file, DGW.xls (previously congapp.xls), is an Excel 97 file containing the data used
in the analysis of Congressional approval presented in Durr, Gilmour, and Wolbrecht (1997).
Those seeking more information on variable measurement or other issues are referred to the
article, or should contact Christina Wolbrecht at Wolbrecht.1@nd.edu or John Gilmour at 
bgilm@malthus.morton.wm.edu.

The variables included in this file are as follows.  Data are quarterly, 1974:1 - 1993:4.

YEAR    
QUARTER 
CAPP    Congressional approval measure, created via Stimson's CALC algorithm.
NEWAPP  Presidential approval
        Measured via the standard Gallup question (average used for
        those quarters with more than one poll).  The original series
        was re-estimated with the effects of economic expectations 
        zeroed out; the resulting series was then subtracted from its
        mean and multiplied by -1 during quarters characterized by
        divided government.  For more information on the measurement of
        this variable, see Durr, Gilmour, and Wolbrecht (1997).
ECONEXP Economic expectations
        The predicted values generated from regressing four measures 
        of the objective economy on the University of Michigan's Index 
        of Consumer Sentiment.  For a more complete discussion of this 
        process and reasoning, see Durr, Gilmour, and Wolbrecht (1997) 
        and Durr (1993).
NYTAVG  New York Times' coverage of Congress
        Variable measures location, length, and spin (postive or negative)
        of coverage of Congress in the New York Times.  Series is an
        average of two series generated by two independent coders.  For
        a more complete description of the coding of this variable, see
        Durr, Gilmour, and Wolbrecht (1997).
NODIV1  Dummy, taking the value 1 in the first quarter of 1977, zeroes 
        elsewhere.  Represents the onset of non-divided government.
NODIV2  Dummy, taking the value 1 in the first quarter of 1993, zeroes 
        elsewhere.  Same as NODIV1.
SC      Series indicating the presence of small Congressional scandals.  
        See Durr, Gilmour, and Wolbrecht (1997), note 9 for a more 
        complete description.
KG      Dummy variable for the Koreagate scandal (1976:4 - 1978:4).
AB      Dummy variable for the ABSCAM scandal (1980:3 - 1982.1).
WR      Dummy variable for the Jim Wright scandal (1989:2).
K5      Dummy variable for the Keating Five scandal (1990:3 - 1991:4).
HB      Dummy variable for the House Bank scandal (1991:3 - 1992:3).
PO      Dummy variable for the Post Office scandal (1992:2 - 1992:4).
VETOES  Count, number of presidential vetoes, including pocket vetoes
        (counted as one-half of regular vetoes).
OVERRIDE Count, number of veto overrides.
INTRASUM Measure of internal congressional conflict.  Count, sum of
        cloture votes and debt ceiling bills.
MBILLS  Count, number of major bills passed both houses.
BILLS   Count, all bills passed by both houses.



REFERENCES:

Robert H. Durr, John B. Gilmour, and Christina Wolbrecht.  1997.
  "Explaining Congressional Approval."  American Journal of 
  Political Science 41:175-207.

Robert H. Durr.  1993.  "What Moves Policy Sentiment?"  American
  Political Science Review 87:158-70.