I agree with what others have written. Your survey assumes that one's reaction to a clip of music will be limited to one of just three possible responses, and that those responses will be limited in descriptive complexity to (1)
happy, (2)
sad, or (3)
scary/suspenseful.
Though I'd describe the fourth and seventh musical examples as turbulent or discordant, the other five were placid and neutral for me. None of them conveyed, or elicited, any of your three predetermined choices.
It occurs to me that a study of
program music, including film soundtracks, would have relevance to your project, as it's explicitly intended to reflect the sentiments of extramusical events (or to serve to stimulate a response to them).
You might also consider that musicians (even across various cultures and musical interests or styles) are not a statistically representative group upon which to base any conclusions. While I doubt that anyone can entirely separate emotions from thoughts, people's background, education and experiences with music can be predicted to have a very significant impact on how they react (and, in turn, how they conceptualize their response).