This paper discusses the influence on my research and writings of several methodological principles that we, the members of the LSE Staff Seminar on Methodology, Measurement and Testing, learned directly from Joseph Agassi and indirectly from Karl Popper. It begins with the origins of the seminar and my textbook, An Introduction to Positive Economics. It goes on to cover methodological issues that arose in my subsequent papers, including: the importance of having empirical content in economic theories, the poverty of theories that are built only to pass sunrise tests, why non-robust assumptions need to be tested, the concept of refutability, the fussy distinction between normative and positive statements, the impossibility of giving purely positive policy advice, the testing of existential statements, fallacious attempts to deduce empirical propositions from definitional identities, the distinction between internally and externally driven research programs, the poverty of modern welfare economics as a guide to policy and the possibility of deriving policy advice without such guidance. It concludes with a short discussion of the revolutionary implications of accepting technological change as being generated endogenously under conditions of genuine uncertainty rather than measurable risk.