There are two senses in which the term "luck" is used, the prescriptive, and the descriptive. Luck, in the descriptive sense, is merely a name we use in describing fortunate events after they have already happened. However, luck in the prescriptive sense is what is meant when one says they either have, or don't have, a belief in luck.
Darke and Freedman (1997) were the first researchers systematically to address directly both the concept and the measurement of belief in luck as a deterministic and personal attribute. They define luck belief as the perception that good luck is ‘a somewhat stable characteristic that consistently favors some people but not others’. They define disbelief in luck as ‘a tendency to agree with the rational view of luck as random and unreliable’ (p. 490). To capture their unidimensional definition of irrational luck belief, Darke and Freedman developed a 12-item measure. Unfortunately, they found their measure ‘does not seem particularly good at distinguishing between people who [say] they [are] typically lucky from those who [say] they [are] typically unlucky’. They also found factor analyses of their measure produced a multi-component solution, as did Prendergast and Thompson (2008).
André (2006) proposed a model of luck-related perceptions that includes separate positive and negative beliefs. However, she found the positive and negative components of personal luck beliefs correlate highly, suggesting they are conceptually very close or in fact the same. Maltby et al. (2008) proposed a 6-dimensional model of beliefs around luck, but empirical analyses supported only a 4-dimensional model: belief in being personally lucky; belief in being personally unlucky; general belief in luck; and rejection of belief in luck.
Thompson and Prendergast (2013) clarified the concepts of belief in luck and belief in personal luckiness. They addressed the logical problem that nobody who disbelieves in luck could consider themselves lucky by differentiating between belief in luck as a deterministic phenomenon that affects the future, on one hand, and on the other, belief in personal luckiness as an appraisal of how fortunately or otherwise chance events in the past might have turned out. They developed and validated an internationally applicable scale to measure, respectively, belief in luck and personal luckiness constructs. They found no correlation between the constructs and no evidence of a distinction between positive and negative aspects of each, suggesting they represent two discrete and unidimensional constructs. Belief in luck and personal luckiness were also found to correlate differently with personality and psychological variables, such as the Big Five and affect.