The quantitative data gleaned from the questionnaire while easy to analyze and compare was limited - no evaluation was made of the participants baseline emotions before starting, and questionnaires provide limited opportunity for elaborating on feelings.
The sample included only men, which reduces the generalisability of the results, as men may experience emotions differently than women.
Not only that, but the experimental conditions were very different than how people feel emotions in real life. They do not usually experience physiological arousal with no warning or explanation. This reduces the external or ecological validity.
Also, epinephrine is known to affect people differently (for example, 5 participants experienced no physiological effects), so this introduces participant variables that may effect the dependent variable.
Participants were deceived about both the aims and procedure of the study as well as the injection they received, making it impossible for them to provide true informed consent.