Situation often arise in statistical modeling where at least one independent variable in the model is imprecisely measured. Statistical models used in such situations are known as measurement error (ME) models and they can be useful in many real world applications. For example, the problem of estimating the association between the time until the diagnosis of leukemia and the level of the exposed radiation dose of survivors of the atomic bombs in Japan is complicated by the fact that a perfectly accurate measure of exposed radiation dose is not possible. Having at least one independent variable measured with error leads to an unidentified model and a bias in the naive estimate of the effect of the variable measured with error. So, for example, ignoring the fact that the measured amount of radiation dosage is not the true amount, would lead to a biased estimate of the effect of radiation on the time to diagnosis leukemia. One method used to correct for his bias and provide information to identify the model is he use of an instrumental variable (IV). An IV is one that is correlated with the unknown, or latent, true variable, but uncorrelated with the measurement error of the unknown truth. So, in our example, an indicator of epilation, or hair loss due to radiation, could serve as an IV since it is correlated with the true amount of radiation dosage, but uncorrelated with the measurement error in estimating radiation dose. We propose a method for parameter estimation, where an IV provides the identifying information, in simple nonlinear ME models; „simple“ because we consider the case where there is only one independent variable in the model and „ME model“ because that variable is measured with error. This class of simple nonlinear ME models includes generalized linear ME models. The initial step in our estimation method is to „categorize“ all continuous and discrete variables. Assuming conditional independence given the latent variable, the joint distribution of the categorized manifest variables equals the product of the conditional categorized distributions summing over the categorized version of the latent variable. Maximum likelihood estimates of the joint categorical distribution are used to solve nonlinear equations for he parameters of interest which enter through the conditional probabilities, which are written as definite integrals or sums of the conditional distributions of the manifest variables. Estimated generalized nonlinear least squares, a method that produces consistent estimates given consistent initial estimates, is used to solve the equations fort he parameters of interest. Wes hall determine the asymptotic properties of our estimators and develop methods of inference for them. We will show how many commonly studied ME models fit into the general framework developed and ccn be solved using our method.