People use values as a really abstract and vague term, when they all seem to hinge on what maximizes human prosperity in a balance with what maximizes individual freedoms.
In that sense, we can look at the effects of various behaviours, and objectively assess whether those effects work in favour or against these things.
Just because our conception of it is vague and personal at this point, doesn't mean that we can't do better in assessing action and consequence.
Yes, science can determine human values, because there are some scientific studies that have revealed that we are far more predictable than we realize. So much of what we do and say and think is controlled by the neurons in our body. Science can study our genetics, which leads to being able to predict our values.
Yes, science can be used to determine human values. We can use science to study when certain morals add to the welfare and well being of humans and other living creatures. We can use this example provided to us by science, along with our common sense, to help us gauge what is moral or immoral.
Science has don the opposite of determining human values by convoluted the issue even more. For example abortion is now expedited by a pill form. Human traits are now proposed to be weeded out by altering genes. Soon the possibility of euthanasia will be elevated by science in an attempt to solve all our problems.,
Science is a great thing but it can not determine the value of a human or human morality. Each person on the planet is worth something. No one is worth more than another just because they are more attractive, have more wealth, more power and more position than someone else. Human values ie morality is something decided by humans as a group.
I do not believe science can determine human values. I believe each person develops their own set of values through life experience and there is no one pre-fixed set of values that every human being has. Secondly, there is no way to scientifically measure these values. Pseudo-science may be able to do it in non-scientific ways, but with that you have problems validating it, as always.