Log in Page Discussion History Go to the site toolbox


From BluWiki

An Introduction to Modes in the Social Sciences Lave and March 1993

Chapters 1-3

By: Megan DeMasters

Chapter 1:

"Speculation is the soul of the social sciences" (p.2) The purpose of this book is to develop a practical guide to speculation and includes components of developing, elaborating, contemplating, testing and revising models of human behavior. Ultimately, this book contributes to our understanding of what a model is.

The authors define a model as a simplified picture of a part of the real world thats purpose is to represent or explain certain phenomenon (p.3). Because models are only a snapshot of the real world it is common to have multiple models that look at some thing. As social scientists we construct models as a way to explain the world around us.

Four necessary skills to model construction:

1. Abstract generalities to the model from the complexities of reality

2. Derivation: within an abstract model be vale to derive significant observations and develop skills to produce meaningful implications

3. Competently evaluate models: discern adequate from inadequate models

4. Have familiarity of common models.

Four common models we should familiarize ourselves with:

1. Individualistic choice: examines the processes by which individuals choose from alternatives, make decisions and solve problems.

2. Exchange: relationship between individual and collective choice

3. Adaptation: how individual and group behaviors change in response to particular events or experiences.

4. Diffusion: spread of behavior, attitudes, knowledge and information through society.

Chapter 2: The authors contend that the best way to learn about model building is to do it (Causal mapping anyone?)

Steps in model building (p.19-20): 1. Observation 2. Examine facts as end-result of model , speculate about the processes that may have influenced the end-result. 3. Deduce other things from the model: implications, consequences, predictions. 4. Ask if other implications are true- reformulate if necessary.

Rules of model building (p.40-42):

1. "Think Process"-good models have a statement about a particular process

2. Develop interesting implications

3. Look for generalities.

Chapter 3:

In evaluating speculations must look at the role of truth, beauty and justice in model building.


Examines the correctness of the model- the authors connoted that the test of truth should be of derivations, not of assumptions:

1. good models can rebuilt on seemingly unreasonable assumptions, we don't want to reject potentially good ideas quickly just because assumptions at first seem unreasonable.

2. Questioning assumptions can be difficult because assumptions may not be directly observable.

3. Looking at assumptions keeps a researcher from questioning the observations of the model, had the potential implications.

Types of Models and Relation to Truth:

1. Circular models: models that can always be correct no matter what happens- this model is bad because it doesn't satisfy the requirement that models are testable.

2. Critical Experiments: compare how various models evaluate the same phenomena. This forces the researcher to determine which model has most explanatory value.

The authors offer several ways to protect oneself from relying too much on one model: (1)think as much as possible about alternate models (2), place less emphasis on seeking truth and (3), be playful in model building.


"Models are art" (p.61). Characteristics of beautiful models

1. Simplicity: the more general the model, or less number of assumptions are ways to increase the simplicity of models. Ways to ensure your model is simplistic are to avoid saying everything you know about a topic, not to worry about counter-examples, and realize that listeners/readers won't develop as much time as you will in assessing your model.

2. Fertility: relatively large number of interesting predictions per assumption

3. Model is unpredictable: implications are surprising to people.

Lave ad March also advise that researchers play to their analytical strengths in developing models. This can include twisting observations to fit a model that you are comfortable with or choosing questions that can be tested with techniques you are comfortable with.


Justice refers to how models contribute to making the world better. This assumes that models are not neutral. The authors recognize that it isn't clear on how we make life better through speculation, and requires researchers to take the offensive to initiate subtle changes. It is very unlikely that we can achieve this, but it doesn't mean that we don't try.

Site Toolbox:

Personal tools
GNU Free Documentation License 1.2
This page was last modified on 19 October 2013, at 00:51.
Disclaimers - About BluWiki