Study an argument or theory to see if it tries to explain the economy or how money works everywhere. Academics call such arguments "theories of everything." Theories of everything are inherently flawed, because they gloss over anomalies that emerge through the interactions of billions of people with free will.
Find any phrasing or arguments that make an absolute claim (using words such as "all," "everyone," or "never"). For example, "everyone uses credit cards" or "all businesses are profitable" are sentences that should raise concern.
Come up with situations in which the theory or argument does not apply to everyone or does not explain everything, and use those situations to prove that the theory or argument is flawed because it does not actually apply to all situations or people.
Find graphs or equations in your economics textbook or an economics article that are not backed up by any evidence or hard data. Nobel-prize winning economist Paul Krugman noted in a 2009 New York Times article entitled "How Did Economists Get it So Wrong?" that economists often fixate on complex equations that have little actual basis in the real world or simple graphics that ignore the complexity of economic interactions.
Ask for evidence or an explanation (that does not rely on mathematics) of how the theory works in the world.
Criticize anything that is too simple as not actually explaining behavior and criticize any lack of hard data (interviews, experiments, even statistics).
Find any personal experience or theories from other disciplines that contradict an economic argument or theory.
Examine a theory or argument (and the evidence or study the theory is based on) for any point where factors were excluded or ignored, or where unrealistic assumptions were made (such as the common one that all individuals are rational about how they spend money).
Ask your professor or the person you are arguing with to explain how the theory addresses contributions from other fields or how the theory can be credible when certain key factors are omitted from the analysis.