The studies on Phillips curves relating the rate of unemployment to the rate of inflation are the results of the search for a reliable tool for forecasting inflation and implementing monetary policy. The original Phillips curve has attracted considerable attention and since then, it has undergone several important changes. Although the evidence of a negative relationship between the rate of change in a money variable and unemployment rate has not been proven, it is still considered to be critical for policymaking and is on researchers' agenda ( Niskaken, 2002 ). The existence of a Phillips curve for Turkey is investigated with the linear and non-linear specifications for the conventional and new Keynesian Phillips curves with inflation expectations and natural rate of unemployment for annual (1980-2001) and semiannual (1988:2-2003:1) data sets. The results indicate no evidence of a Phillips curve for all specifications and both data sets. The semiannual data seems to have a better fit in all specifications compared to the annual data. Inflation expectations are found to be significant for inflation rather than unemployment rate in the current period. The results imply that Turkey has to solve its inflation problem by proper policies aimed at lowering the inflation expectations in the economy.