Judy writes a survey to assess how much people worry. Her survey has 20 questions that people can rate their level of agreement to on a seven-point Likert scale. All of the questions are worded so that higher responses will indicate a higher level of worry. After having 100 people complete her survey, she finds that a lot of respondents often respond to all of the questions by choosing only the "strongly agree" option. How could she improve the construct validity of her survey?