Dear Raija-Leena Punamäki, dear discussion group,
I would like to respond to the author's responses.
First, I would like to discuss the problem of eliciting dream recall
frequency via dream diary. Previous research has shown that participation in
a dream study or even a single emphasis made by the experimenter could
increase dream recall in a substantial way. Keeping this in mind it seems
plausible that many factors could contribute to dream recall frequency. What
I meant by the statement 'The participants know the aim of the study' is the
following. Participants have concluded (I think) that the purpose of the
study is to look for relationships between trauma and dreaming (They
received a dream diary ans specialized questionnaires). So it seems
plausible that the high exposure group will focuse on dreams which may be
related to traumtic events. All participants were highly motivated and
reported up to 4 dreams per week (Gaza group). This may be explained by the
focus on trauma-related dreams since adults (students) recall on the average
one to two dreams per week.
Second, I don't know whether you understand my question concerning the
sampling method. You wrote "in one household all children who wanted to
participate... (p. 241). This mean to me that some children in the given age
range participated and some not. I would like to know the percentage of
children who refused to participate.
Third, in my research I used measures of central tendency according to the
measurement level of the dream content scales, i. e., mean, median or modus.
First of all, I will check whether there are substantial correlations
between dream length and the waking-life variables since these relationships
could lead to overestimations of the correlations between dream content
scales and waking-life measures. Bill Domhoff would propose the usage of
ratios in order to control for dream length and number of dreams. One can
compute, for example, the ratio of aggressive interactions per dream
character or the ratio of aggression directed at the dreamer and aggression
directed on others and so on. For more details see Domhoff (1996). Finding
meaning in dreams.
Forth, I would like to know how you computed the correlation between dream
atmosphere/feeling scores and morning mood rating. As I understand you could
only include subjects with at least two dreams to compute correlations. (By
the way, we found also a stronger carry-over effect for positive emotions
that for negative emotions in a small sample of adults).
Fifth, your psychological symptoms scale includes an item (or two items)
concerning nightmare frequency and bad dreams. Did you analyze this (these)
item(s) which seems promising since there is s substantial difference
between the Gaza group and the Galiliee group (p. 247). I would compute an
analysis as previously mentioned for the number of negatively toned dreams
in that one-week period to test the continuity hypothesis of dreaming. It
seems not plausible to me that the traumatic environment increase dream
recall frequency per se but recall of negatively toned dreams. These
analyses will add much material to the discussion of the present results.
Sixth, I do not understand why you have not carried out a regression
analysis including all variables, i. e., salience measures, age, gender,
trauma-related measures, coping styles, and the three subscores of the
Overall, I think the interpretation of the data is not an easy task. As I
mentioned earlier the explained variance of the regression analyses is about
10 % (except for the analysis including salience measures). This should lead
to very careful interpretations since a variety of other factors may
contribute to the obtained results.
Dipl. Psych. Michael Schredl
Central Institute of Mental Health
P. O. Box 12 21 20
Telefon: Germany 0621/1703-602
Fax: Germany 0621/23429