Architecture Research Methods ARCH 5365 G

1
Architecture Research MethodsARCH 5365GW
Chapter 8

• Gary W. Smith, 2006

2
Correlational Research

• Definition clarifies patterns of relationships
  between two or more variables, i.e. factors
  involved in the circumstances under study.

3
General Characteristics

• A focus on naturally occurring patternscomplex
  sets of real-world variables (the range of
  characteristics that vary with the setting being
  studied and that influence socio-physical
  interaction).

4
General Characteristics (Cont.)

• The measurement of specific variablesmust be
  able to be quantified and measured. Can involve
  observation, surveys/questionnaires, and physical
  measurements.

5

• TYPES OF MEASUREMENTS
• Categorical Measurement the variable of interest
  is sorted into discrete categories, based on
  verbal or nominal terms.
• Ordinal Scales more precise than categorical.
  Rankings, i.e. 1 to 10 on a scale.
• Interval and Ratio Scales Still more precise.
  Specifies the exact distance (or interval)
  between one measurement and another. Interval
  relies on an established and consistent unit of
  measurement i.e. dollars, feet, degrees, etc.
  Ratio scales are those that start with 0 where
  one can state that one measurement is x times
  larger than another.

6
General Characteristics (Cont.)

• the use of statistics to clarify patterns of
  relationshipsgraphic charts, correlational
  statistics that describe the strength and
  direction of a relationship between two or more
  variables. (Scientific method)

7

• Two types of correlational research
• Relationship Studiesfocuses on the nature and
  predictive power of relationships
• Causal-Comparative Studiesare between
  relationship studies and causality that
  characterizes experimental research. These
  studies select COMPARABLE groups of people or
  COMPARABLE physical environments and then collect
  data on a variety of relevant variables.
• Can only ascribe cause in a provisional or
  hypothetical way.
• Cannot guarantee that the reaction to variables
  is totally due to those variables.

8

• What is sampling?
• determining the number of people to include and
  which ones.
• Important because the goal of correlational
  research is to predict as accurately as possible
  the response or behavior of a large group of
  people, based on the patterns established by a
  smaller subset of that group.
• Probabilistic sample to achieve a sample that is
  representative of the larger population. Usually
  random.
• Nonprobabilistic or purposive sample less about
  generalizing to the larger population and more
  concerned with discovering patterns of
  information about a particular group(s), i.e.
  programming a building.

9

• Tactics for collecting data
• Surveys
• Observation
• Mapping
• Sorting
• Archives

10

• Surveys
• can provide a lot of information in a short
  period of time, but may not provide an in-depth
  understanding of the issues surveyed. In-depth
  understanding is better in qualitative research

11

• 2. Observation
• Must first decide what to look for
• Where to look for it
• When to look for it

12

• Mapping
• Putting results into a graphic form

13

• 4. Sorting
• Directed sorting involves putting things into a
  predetermined set of categories
• Free sort letting the subject put thins into
  whatever categories they want
• Multiple free sort doing free sorts multiple
  time with different categories.

14

• 5. Archives
• Using existing databases to get information.

15

• Three types of analysis methods
• Multiple regression
• Factor analysis
• Multidimensional Scaling

16

• Strengths
• Can clarify the relationships among two or more
  naturally occurring variables
• Well suited to studying the breadth of a setting
  or a phenomenon
• Can establish predictive relationships

17

• Weaknesses
• Researcher cannot control the levels or degrees
  of variables (this would be done in experimental
  research)
• Less well suited to exploring the setting or
  phenomenon in depth
• Cannot establish causality

18
Notes

• Information in these lectures was adapted from
  the following sources
• Booth, Wayne C., Colomb, Gregory G., Williams,
  Joseph M. 2003. The Craft of Research. Second
  Edition. Chicago University of Chicago Press.
• Groat, Linda N. and Wang, David C. 2002.
  Architectural Research Methods. New York John
  Wiley.