When can I know that I'm in the presence of measurement errors? i.e., when can I know that I should use a measurement error model(error-in-variables model)?
Are there any statistical tests for checking that?
When can I know that I'm in the presence of measurement errors? i.e., when can I know that I should use a measurement error model(error-in-variables model)?
Are there any statistical tests for checking that?
Suppose we observe only $\tilde{x}$ and $\tilde{y}$ which are the true values measured with error: $$ \tilde{x} = x + u$$ $$ \tilde{y} = y + v.$$ We would like to estimate: $$ y = \beta x + \epsilon$$ but all we can really estimate is $$\tilde{y} = \hat{\beta} \tilde{x} + \zeta$$
However, if y is measured without error ($\tilde{y}=y$) we can use reverse regression to bound the measurement error in regressions. The regression is:
$$ y = \hat{\beta} \tilde{x} + \epsilon$$
You also run the reverse regression $$ \tilde{x} = \hat{\frac{1}{\beta}}y + \hat{\frac{1}{\beta}}\epsilon+ u$$ Define the reverse regression beta as $\hat{\beta}_r = \frac{1}{\hat{\frac{1}{\beta}}}$. It turns out that the results from the standard regression and from the reverse regression will bracket the true coeffcient, i.e. $plim\ \hat{\beta} \leq \beta \leq plim\ \hat{\beta}_r$.
This interval is small when the variance of $u$ and $\epsilon$ are small. If one of the two are zero then one of the inequalities hold exactly. Unfortunately, in practice this interval may be large and therefore this approach not very helpful.
With observational data, I would say measurement error always exists.
The real issue is "how severe" is the measurement error. Evidently in order to be able to assess the severity, "experimental-like" (or quasi-experimental) data should be available.
For an example, see this paper
Boyd, D., Lankford, H., Loeb, S., & Wyckoff, J. (2013). Measuring test measurement error: A general approach. Journal of Educational and Behavioral Statistics, 38(6), 629-663.
here downloadable freely from its initial NBER version.
They treat test-scores essentially as a proxy for "student achievement" (knowledge, skills and abilities), so we are not talking about errors in grading the tests. As the authors write
"Reliability coefficients based on the test-retest approach using parallel test forms is recognized in the psychometric literature as the gold standard for quantifying measurement error from all sources. Students take alternative, but parallel (i.e., interchangeable), tests on two or more occurrences sufficiently separated in time so as to allow for the “random variation within each individual in health, motivation, mental efficiency, concentration, forgetfulness, carelessness, subjectivity or impulsiveness in response and luck in random guessing” but sufficiently close in time that the knowledge, skills and abilities of individuals taking the tests are unchanged."
Testing for measurement error in linear regressions: Hausman (1978, Econometrica, www.jstor.org/stable/1913827)
Testing for measurement error in general nonlinear, nonparametric models: Wilhelm (2018, CeMMAP Working Paper, http://www.ucl.ac.uk/~uctpdwi/papers/cwp451818.pdf)