Contingent valuation (CV), choice experiment (CE) and experimental auction (EA) or the combinations of the three methods are often used by researchers to elicit consumer willingness to pay for food attributes (food labels). One concern about using these approaches is that quality attributes of food provided to respondents are assumed independent of other attributes which are not provided to respondents during the survey. The limited attributes provided in a survey may lead respondents to allocate their budgets to those limited attributes rather than allocate their budgets to a larger number of product attributes to truly reveal their preferences. Surveys containing a series of online CEs were collected to investigate the effects of additional beef steak attributes on consumer WTP in two different US markets. Random parameters logit models are estimated for each CE in the questionnaires with survey results from both samples. The models with the different survey samples reveal consistent results regarding changes in WTP with more attributes added to the CEs. Consumer WTP for the most important attributes in the CE decreases when the number of attributes increases from three to four, while the WTP for the most important attributes increases when the number of attribute increase from four to five. The changes in the WTP for attributes depend on their relationships with the newly added attributes to the CEs and the number of attributes in CEs.