Architecture Research Methods ARCH 5365 G - PowerPoint PPT Presentation

1 / 18
About This Presentation
Title:

Architecture Research Methods ARCH 5365 G

Description:

A focus on naturally occurring patterns complex sets of real-world variables ... Can only ascribe cause in a provisional or hypothetical way. ... – PowerPoint PPT presentation

Number of Views:835
Avg rating:3.0/5.0
Slides: 19
Provided by: larry262
Category:

less

Transcript and Presenter's Notes

Title: 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.
Write a Comment
User Comments (0)
About PowerShow.com