The marker variable technique offers a practical and easy method to address the validity threat on account of common method variance to the findings of individual studies. However, the validity of the marker variable technique has not been established. This paper examines the validity threat to the marker variable technique arising from multiple criteria for the selection of the marker variable proposed in the literature. Our findings show that the conclusions that can be drawn under different criteria are significantly and substantially different, raising doubts about the validity of the marker variable technique. Proposals for developing theoretically grounded criteria for identifying the marker variable correlation are discussed. Importantly, our analysis suggests that the second lowest correlation of the criterion variable may be a more valid method for selecting the marker variable correlation.