Content analysis on the potential significance of color in dreams: A
preliminary investigation

Methods and Results2.

Content analyses were performed on two large dream databases. The studies
looked for common patterns in the relative frequencies of each reported
color. The content analyses were performed on a combination of two
databases together, totaling 30,063 dreams. One database used in the
analyses included 25,222 dream reports from the DreamBank.net database
(Schneider & Domhoff, 1999), which consisted of a large population of male
and female dreamers of varied demographics primarily from the US, but also
samples from Europe and South America. This original analysis was updated
in February of 2010 for the purposes of this report. The second database
included 12,841 dream reports from a small male and female population of
eight long-term journaling subjects from the US and UK. A content analysis
was performed on the total combination of dreams from both databases, as
well as each individually. Also, each of the eight long-term journal data
sets were analyzed in order to study individual differences, in comparison
with each other, as well as to the composite results from the whole
population.

Color Naming and Search Criteria2.1.

One of the more difficult variables in such a content analysis search is
the naming of colors, which can vary by culture, gender and personal
experience with color (artists for example may have more names for subtle
color hues and combinations than others). An initial examination of the
collected journaling data, plus trial naming searches on the DreamBank.net
database, indicated that identification of colors with a “minor” term (for
example “scarlet”, “saffron”, “hazel” or “flesh”) occurred very
infrequently (typically much less than 1%) in comparison to the use of more
primary terms. It may have improved the accuracy a percent or so to try to
collect these minor terms into the closest primary color, however this
could also add a subjective variable where the match was unclear. It was
decided to proceed with sorting on primary color naming since the
occurrence of any one minor color term across a large database was so
small. Yellow and purple were two exceptions where a “minor” name was
combined into the count, due to the higher frequency of occurrence and the
common tendency to use alternate names based on the context. Yellow counts
included “blond” since it was most often used when dreamers were describing
yellow hair (Western speech being more inclined to use the term “blond”
than yellow when naming hair color). “Golden” was often used to describe a
yellow glow but was eliminated from the yellow count since it shared
characteristics with the term “gold”. “Violet” was included in Purple
counts since this was a commonly interchanged word for the same color.

The second problem was color combinations, for example “teal” as a
blue-green combination, or “tan” as a yellow-brown combination. With minor
exceptions, such as “pink” (which occurred about as frequently as orange)
these combination names also typically occurred much less than 1% of the
time.

Although the sum of all the minor color terms could add up to a noticeable
percent in some individual reports, the study was not focused on total
color count but rather dominant discrete colors which form patterns that
appear well above the levels of “minor” color names. The low levels of
occurrence provided confidence that looking for color patterns using more
primary color names was reasonable. A search of both databases resulted in
the dominant color names to be: red, orange, yellow, green, blue, purple,
brown, gray, black, white, gold, silver and pink. This set of dominant
colors was used to compare color recall rates. Although it is not inclusive
of all colors (minor terms, mixed colors, “rainbow” colors, and
non-specific colors) the purpose was not to establish a total color recall
rate in this paper but to compare recall for a fixed set of colors. Silver,
gold and pink were dropped from the graphs herein since the intent was to
look for patterns and the occurrence of those minor colors was low and the
explicit color identification less discrete. The remaining colors are
termed the “standard color set.” Word count was therefore plotted and
compared for the following color terms: red, orange, yellow/blond, green,
blue, purple/violet, brown, gray, black and white.

Composite Database Trial2.2.
The dream reports in both the Long Term Journaling and DreamBank.net
databases were searched for discrete color words using the “standard color
set.” The combined search of 30,063 dream reports resulted in a count of
12,227 for these color terms. The relative frequency between the colors
reported is illustrated in Figure 1 as a percentage of the “standard color
set.” The analysis found that on average black and white (as colors) are
reported most often (approximately 20% each) and with approximately equal
frequency. This is followed by the next dominant frequency grouping of the
colors red, blue, yellow and green. Within this grouping, red tended to
appear a larger percentage of the time (15%) in comparison to the other
three colors (which occur about 10% of the time). Other colors were
reported less than 5% of the time with the possible exception of brown,
which was reported in the 6-7% range. This color pattern is significant,
since it is observed to exist in some form in a majority of the large and
many small data sets.

Conclusions3.

A fairly common pattern seems to emerge when dream color recall is studied
across large populations, as well as within individual dream journals. This
pattern, which I will term the “dominant pattern” herein, consists of a
dominant pairing of black and white (named as colors) appearing with
approximately equal frequency, followed by a grouping of the “primary” hues
red, yellow, green and blue (with red appearing about 50% more frequently
than the others). This grouping is followed by brown which often appears
twice as frequently as the lesser colors. This pattern appears fairly
consistent between the two databases, and appears as well in the journals
of individual dreamers, but with a wider variation between the relative
color counts between individuals. These results lead to the speculation
that there may be a common neurological or psychological factor influencing
dream color creation or recall on average, which is in turn influenced by
other factors at an individual level. The discussion below compares the
data to each hypothesis in order to determine if it strengthens or weakens
each hypothesis as a contributing factor.               KR   IRS      19 6
23

-- 
You received this message because you are subscribed to the Google Groups 
"Thatha_Patty" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To view this discussion on the web visit 
https://groups.google.com/d/msgid/thatha_patty/CAL5XZopwh8kW-Xfwfy9BiD%2BFmHJUjezcXF7OrKzMi98r-ERHKQ%40mail.gmail.com.

Reply via email to