I just read a NYT article The Data Against Kant that challenged the philosopher Emanuel Kant's thinking about ought verses can. They gave an example where a friend and you were interviewing for the same job and she offered to drive you to the airport, but the car broke down. Does she still have an obligation to get you to the airport? Most people said "No" until they changed the scenario to one where she purposefully sabotaged the car to stop you getting to the airport.
The problem comes when you feel that she sabotaged your chances rather than it was an act of fate. The difficulty with the example is the same difficulty with most research. Research tends to lump what happens in the environment with relations and even connect events to a biological cause without trying to understand the interactions of the parts. This basic problem is not paying attention to what you are researching.
Which area is it - environment, relationship, or biology? If the researcher differentiates as much as possible and then tries to define the interconnect, the research makes more sense. For example, if a car breaks down it is an environmental event that does not affect your relationship. It may cause a disappointment, but doesn't damage the relationship. If the person sabotaged the car, it is not an environmental issue, but the relationship is harmed. Finally, if the person has a heart attack, which is a biological event, the relationship remains in tact. Clarifying the real focus of research makes a huge difference in the outcome.