In this paper, we aim to measure agricultural productivity change of 34 OECD countries between years 1990 and 2014. The methods employed are data envelopment analysis (DEA) and malmquist productivity index (MPI). DEA is a relative efficiency method in a production technology, whereas MPI is based on DEA to measure the changes in the production technology over time. Our challenge is the existence of missing data points over the years in the initial dataset, which correspond to approximately 9% of the data. Removing units, factors or years with missing data as commonly practiced in DEA, would cause loss of information and makes it very difficult to draw conclusions in such a macro-data. We present the idea of using averages of available data points for a given factor and average variations over the years in those data to produce intervals for the missing points and handle the problem without any dimension reduction in the available data. Fuzzy DEA approach is employed using the calculated factor-specific intervals followed by MPI calculations to conduct a productivity change analysis. We suggest and empirically illustrate that instead of narrowing down the scope of the analysis by excluding the points missing, applying fuzzy approaches is an option worth considering by which it can be possible to make the best out of the available information. The results of the analysis are interpreted with respect to years, countries, regions and economic size of the countries.