Economic Sociology
1
,Chapter 1: Sociology’s Status
1. Science and pseudo-science
• Science has had impressive success in understanding and controlling material and
biological reality
• Central question: what distinguishes successful science from dead ends and everyday
judgments?
• Dead ends: sometimes denotes as ‘pseudo-science’ (astrology, homoeopathy)
• In the long term, at least four (connected & partially overlapping) elements have proven
quintessential
• Learns from its mistakes, leaves room for improvement
a. Internal consistency
• Logical errors, contradictions, conceptual clarity, openness to falsification
o Language and structure of arguments (mathematic language is conducive to
consistency)
b. Evidence
• Good theory is necessary condition, but also needs confrontation with data, ‘facts’
• In natural science: experiments
o The scientific cycle consists of a and b in continual confrontation (theory and
facts)
c. Reduction of error
• Some speak of ‘progress’
• However, this wrongly suggests a premeditated direction
• Essential: it is about the direction of reduction of errors, not about making no mistakes
(to the contrary!)
• On average, it is quite safe to say that in science errors become less frequent and less
fundamental
• Very important implication: there is no validity in ‘the age of an idea’, nor in thw opposite
• An indicator of pseudo-science: ‘it is new/ancient knowledge, therefore it is valid’
d. Social organisation
• Central element underlying a-c
o Individual (ethical, attitudinal) dimension: be receptive to refutation, even seek it
(acquired socially)
o Social: it presupposes a system of competition and free exchange of ideas, with
some parallels to market exchange (and some fundamental differences!)
▪ Case: Lysenko, Russian agriculturalist, claimed that plants could inherit
acquired characteristics. Officially sanctioned as Soviet policy by Stalin,
leading to expulsion and imprisonment of biologists/agricultural
scientists
2
,2. Can sociology be scientific?
• In all four dimensions, sociology can meet ‘scientific’ standards (but, as every science, it
can turn pseudo-scientific)
• However, at least two fundamental aspects make sociology differ fundamentally from the
natural sciences
o Sociology can hardly build on experiments
o Sociology deals with humans and their interactions
a. The experimental design
• An experiment is historically the oldest technique to come to scientific judgments
• It is also the purest, most certain way to arrive at valid and reliable conclusions
• What is an experiment (in its pure form): OXO
• This OXO model boils down to 3 elements
o Randomised distribution in 2 groups (an experimental versus a control group)
o Former gets a treatment, latter not
o Observation (measurement) before and after stimulus
• More than alternative ways of testing theories, OXO allows for reliable and valid
conclusions
• This makes experiments the most ideal way of testing causal relationships (not
necessarily other goals of scientific knowledge)
• Ground of its superiority: isolation of the effect of treatment (no need to know other
relevant factors)
• Later on, variants have been developed, e.g. for controlling for causal bias (Solomon,
placebo)
• Sociology deals with objects that are hard to imitate in laboratory contexts
o The Northern Irish civil war
• Second best options probably are (large-scale) surveys and quasi-experiments
• Alternative 1: Large scale surveys
o Usually based on (standardised) interviews
o Many variables collected
o Downside in relation to experiments: no control
▪ No certainty about causal status of correlates (spuriousness)
o Ex. 1: ice cream and drowning at the seaside
o Ex. 2: listening to ‘unconventional music’ and deviant activities (ter Bogt et al,
2013)
o Dominant strategies: control for possible intervening variables
▪ Multivariate designs
o Watch the same elementary units (persons, households, nations) change over
time
▪ Time series and longitudinal designs
• Alternative 2: Quasi-experiments
o Often: experiments in the real world (field experiment), so with diminished
control
o Also: non-randomised distribution of groups
o Most promising: natural experiments: ‘accidental’ experiment in real world
▪ Cunningham’s ‘men in transit’
3
, o Comparison with survey
▪ No intervention by the researcher
▪ Mimics experimental design
b. Human Action
• The object of sociology is: “conscious sentient beings”
• Action of people is fundamentally different from logic of ‘laws’ in nature
• Illustration: water boiling and people in a sauna
• Basically, explanations in sociology are: accounting for reasons (and therefore beliefs,
values, motives, intentions)
• Sociology does less but also more than natural sciences
o Railway station
• Basic strategy: ask people
c. Behaviour, action and social action
• Sociology is the science of social phenomena, made up of(social) actions
• Criterion 1: intentionally
• Criterion 2: coordination
• Beware! The same behaviour belongs to different partial sets
• Types of (social) action
o Instrumental rational
o Value rational
o Affectual (emotional)
o Traditional (habitual)
• Conclusion
o ‘Aimless’ or ‘meaningless’ action is a contradiction in terms
o Non-rational action can also be meaningful
o There is more than one type of rationality
Learning Aid – Chapter 1:
• Science
• Scientific cycle
• Induction
• Deduction
• Reduction of error
• Experiment*
• Quasi-experiment
• OXO-model
• Experimental/control group
• Solomon experiment
• Placebo
• Spurious correlation*
• Action*
• Traditional action
4