P1: GCP/GYQ

Dreaming [dream] ph093-drem-366672 January 5, 2002 11:12 Style file version Nov. 19th, 1999

Dreaming, Vol. 12, No. 1, March 2002 (°

C 2002)

Questionnaires and Diaries as Research Instruments

in Dream Research: Methodological Issues
Michael Schredl1,2

Dream questionnaires are widely used in dream research to measure dream recall frequency
and various aspects of dream life. The present study has investigated the intercorrelation
between questionnaire and diary measures. 285 participants completed a dream question-
naire and kept a dream diary over a two-week period. Results indicate that keeping a
dream diary increased dream recall in low and medium dream recallers but decreased
dream recall in high dream recallers. The correlation coefficients between questionnaire
items measuring aspects of dream content and diary data were large, except for a more
complex scale (realism/bizarreness). In the low recall group, however, considerably lower
coefficients were found indicating that recall and sampling processes affect the response
to global items measuring dream content. Using the example of testing gender differences,
the findings of the present study clearly indicate that the measurement technique affects
the results. Whereas sufficient internal consistency and retest reliability have been demon-
strated for various dream questionnaires, future research should focus on the aspects of
validity by comparing questionnaire data to dream content analysis of at least 20 dreams per
person.
KEY WORDS: dream questionnaire; dream diary; dream recall frequency; dream content; reliability; validity.

Three measurement techniques are widely used to assess dream recall frequency ques-
tionnaires, dream diaries and laboratory awakenings (cf. Schredl, 1999). Questionnaires are
easy to administer though faulty memory may bias answer to such items as “How many
dreams do you have recalled during the last month?” Dream diaries or laboratory awak-
enings minimize incorrect or inaccurate recall but, unlike questionnaires, recall may be
affected by the measurement technique.
Participants pay attention to their dreams while keeping a dream diary, and several stud-
ies (e.g., Cohen, 1969; Reed, 1978; Schredl, 1991) have shown that keeping a dream diary
can increase dream recall frequency markedly. The same is true for laboratory awakenings
(e.g. Goodenough et al. 1959; Cohen & MacNeilage, 1974; Moffitt et al., 1982): Although
the recall rate after REM awakenings of low dream recallers (questionnaire measure) is

1 Central
Institute of Mental Health, Mannheim, Germany.
2 Correspondence should be directed to M. Schredl, Ph.D., Sleep Laboratory, Central Institute of Mental Health,
PO Box 12 21 20, 68072 Mannheim, Germany; e-mail: Schredl@as200.zi-mannheim.de.

17
1053-0797/02/0300-0017/1 °
C 2002 Association for the Study of Dreams
P1: GCP/GYQ
Dreaming [dream] ph093-drem-366672 January 5, 2002 11:12 Style file version Nov. 19th, 1999

18 Schredl

well below that of high dream recallers (e.g. 75% vs. 96%; Cohen & MacNeilage, 1974),
the overall recall rate is dramatically increased in comparison to home dream recall. The
intercorrelation between questionnaire items and diary data is often high (Baekeland, 1970;
Belcher, Bone & Montgomery, 1972; Bernstein & Belicki, 1995–96; Schredl, 1998a).
Cohen (1969), however, reported a null correlation between dream recall measured
by questionnaire and by dream diary. The author argued that low recallers tend to increase
their recall frequency (effect of focusing on dreams), whereas for high recallers a further
augmentation is not possible (ceiling effect). This hypothesis was supported by the findings
of Cory et al. (1975).
Despite the high correlation coefficients between the different measures of dream
recall frequency, the question whether the measurement technique affects the magnitude
of relationships to other variables such as trait aspects, cognitive abilities, sleep behavior
etc. has not been investigated in a systematic way. Early (1977), for example, reported
that introverts recall their dreams more often than extraverts if dream recall was measured
by questionnaire, but these two groups did not differ on the number of dreams reported
in a dream diary. This example emphasizes the importance to take a careful look on the
measurement technique which is utilized and the difficulty in deciding which finding is
more valid since both techniques (questionnaire and dream diary) have their pros and cons
(see above).
The present study was designed to investigate in detail the relationship between dream
recall frequency measured by questionnaire and by dream diaries, i.e. the intercorrelation
of these two measures and the pattern of high and low dream recallers. In addition, gender
differences were studied in order to test whether the measurement technique affects the
result. Several large-scaled surveys (e.g. Borbely, 1984; Giambra, Jung & Grodsky, 1996;
Schredl, 2000) using questionnaire measures have shown that women tend to recall their
dreams more often then men.
Questionnaires were not only utilized for measuring dream recall frequency but various
aspects of dream content (e.g., Domino, 1982; Gruber, 1988; Bernstein & Roberts, 1995;
Kahlmeyer & Chang, 1997; Kroth et al., 1999; Zadra & Nielsen, 1999; Agargün et al., 1999;
Bruni et al., 1999). First, the questionnaire approach is simple and easily applicable to a
large sample and, second, Bernstein, Belicki and Gonzales (1995) argued that questionnaire
items are more valid measures for trait aspects of dream life than content analysis of diary
dreams or dreams obtained from laboratory awakenings. Indeed, the studies of Kramer and
Roth (1979) and Schredl (1998b) have demonstrated that a large number of dream reports
per participant (about 20; Schredl, 1998b) is necessary to obtain reliable respectively stable
measures.
The question arises, on the other hand, what factors influence the answer to an item
concerning dream content (e.g. Do your dreams generally take place: A. Indoors in a familiar
setting; B. Indoors in an unfamiliar setting; C. Outdoors in a familiar setting; D. Outdoors
in an unfamiliar setting; Bernstein & Roberts, 1995). Bernstein and Roberts (1995) have
stressed the person’s self-concept as influencing variable, but it seems plausible that dream
recall frequency, memory for recalled dreams and intensity of dreams are important for
estimating the general aspects of one’s dreams. Whereas data were provided for the inter-
nal consistency of dream questionnaires (e.g. Kahlmeyer & Chang, 1997; Agargün et al.,
1999) and the retest reliability (e.g. Soper et al., 1990; Bernstein & Belicki, 1995–96;
Kroth, Yoneda & Hammond, 1998; Bruni et al., 1999), only one study (Bernstein, Belicki
P1: GCP/GYQ
Dreaming [dream] ph093-drem-366672 January 5, 2002 11:12 Style file version Nov. 19th, 1999

Questionnaires and Diaries as Research Instruments in Dream Research 19

& Gonzales, 1995) has directly compared dream questionnaire data with dream content
analysis results of diary dreams. The correlation coefficients between the corresponding
scales reported in this study were small (up to r = .30) and seldom significant despite the
sufficient sample size (N = 106). The authors argued that this lack of relationship might
be explained by the low stability of dream content; an explanation which is supported
by the study of Schredl (1998b), since five dreams per participant which were used by
Bernstein, Belicki and Gonzales (1995) are not sufficient to measure all aspects of dreams
reliable.
Gruber (1988) has excluded 161 low recallers from data analysis stating: “Due to
difficulties low recallers might have encountered when responding to the DSQ, potentially
resulting in distortion of results, these subjects were removed from the data analysis pro-
cedures (Gruber, 1988; p. 75).” However, he did not test this hypothesis by comparing the
questionnaire data of low and high dream recallers.
To summarize, research indicate that dream questionnaires are reliable measurement
instruments, but validation studies in respect to dream content analysis of dream reports are
scarce.
The present study investigates the relationship between selected measures of dream
content (and self-ratings) derived from questionnaires vs. dream diaries, in particular, emo-
tional intensity, emotional tone, realism, creative aspects of dreams. It also examines whether
dream recall frequency mediates this relationship.

METHOD

Participants

The sample included 285 persons whose mean age was 24.8 ± 9.1 years. Almost all
participants were psychology students in their first year of university studies; five were
pupils and fifteen employees. 266 persons reported at least one dream. Overall, 979 dreams
were recorded. The mean dream length was 114.7 ± 93.0 words.

Research Instruments

Dream Questionnaire

The dream questionnaire is comprised of items measuring dream recall frequency

and some aspects of dream content. Dream recall frequency over the last few months was
measured by a seven-point scale (0 = never, 1 = less than once a month, 2 = about once a
month, 3 = twice or three times a month, 4 = about once a week, 5 = several times a week
and 6 = almost every morning.) Additionally, items were constructed to measure dream
realism (0 = non-realistic dreams, 1 = dreams are sometimes realistic, 2 = dreams are
often realistic), creative aspects of dreams (five-point scale: 1 = no creative aspects to 5 =
many creative aspects), emotional intensity of dreams in general (five-point scale: 1 = not
intense to 5 = very intense) and the overall emotional tone of dreams (−1 = predominantly
negative, 0 = balanced, +1 = predominantly positive). A recoding for the dream recall
scale was done in the following way in order to obtain estimates for the frequency over
P1: GCP/GYQ
Dreaming [dream] ph093-drem-366672 January 5, 2002 11:12 Style file version Nov. 19th, 1999

20 Schredl

a two-week period: 0 = 0, 1 = 0.25, 2 = 0.5, 3 = 1.3, 4 = 2.0, 5 = 7, 6 = 13 mornings

with dream recall per two weeks (see section “dream diary” for the diary dream recall
measure).

Dream Diary

Each participant kept a dream diary over a two-week period. In addition to a checklist
measuring dream recall, participants were instructed to record their dream(s) as completely
as possible and to use two four-point scales to rate the intensity of positive and negative
dream emotions (0 = no emotions to 3 = strong emotions). For both positive and negative
dream emotions, median scores were calculated for the first five dreams recalled, although
when fewer than five dreams were recalled the median score was calculated on the number of
dreams available from that participant. The overall intensity of dream emotion was indicated
by the sum of the medians for positive and negative dream emotions, whereas emotional
tone was determined as difference between positive and negative emotions. Mean word
count was derived by averaging the word count on (up to) the first five dreams. Diary dream
recall was defined as number of mornings with recall of an explicit dream. In addition,
participants could check a category “dreamt, but no recollection of specific dream content.”
A second measure of dream recall was derived by summing the number of mornings with
explicit recall and content-less recall.

Dream Content Analysis

Several rating scales used in the present study were adopted from Schredl, Schröder,
and Löw (1996) and Schredl et al. (1999): realism/bizarreness (1 = realistic to 4 = sev-
eral bizarre associations); positive emotions, negative emotions, emotional intensity and
emotional tone (using the same format as the self-rating scales in the dream diary). The
interrater reliability coefficients (Spearman rank correlations) for these scales are high and
range from 0.70 (realism/bizarreness) to 0.90 (negative emotions) (e.g. Schredl, Schröder
& Löw, 1996; Schredl et al., 1999).

Procedure

Participants were recruited on the campus or from the author’s work setting. Partici-
pation was voluntary and unpaid. The participants were tested in groups (up to 20 persons).
First, participants completed the Dream Questionnaire. Second, instructions about keeping
the dream diary were given orally. These instructions were also included in the dream diary.
The dream diary was kept over a two-week period without any further contact with the ex-
perimenter. Third, dream reports were typed, randomly arranged, and scored on the dream
rating scales by a judge blind to the identity of the dreamers.
To obtain individual dream content measures, means across each participant’s reported
dreams were calculated for word count; medians were calculated for realism/bizarreness,
positive emotions, negative emotions, emotional intensity and emotional tone. Statistical
analysis were carried out using the software SAS for Windows 6.12.
P1: GCP/GYQ
Dreaming [dream] ph093-drem-366672 January 5, 2002 11:12 Style file version Nov. 19th, 1999

Questionnaires and Diaries as Research Instruments in Dream Research 21

Table I. Dream Recall (Questionnaire (Q) and Diary)

Recall group Dream recall (Q) recoded Dream recall (Diary) t-test

Low recall (1, 2) (N = 35) 0.39 ± 0.13 2.03 ± 1.87 t = 5.7 .0001
Medium recall (3, 4) (N = 124) 1.74 ± 0.35 3.14 ± 1.97 t = 7.8 .0001
High recall (5, 6) (N = 126) 8.62 ± 2.67 5.53 ± 2.82 t = −11.2 .0001
Total sample (N = 285) 4.62 ± 4.01 4.06 ± 2.73

All probabilities are two-tailed.

RESULTS

Dream Recall Frequency Measures

The correlation coefficient between dream recall (questionnaire) and dream recall
(diary) was r = .557 ( p < .0001). If the sample was divided according to the question-
naire dream recall frequency into three subgroups (low recallers = about once a month
or less; medium recallers = several times a month to about once a week and high re-
callers = several times a week to almost every morning), the comparison between ques-
tionnaire and diary data revealed that low and medium dream recallers tended to increase
their dream recall frequency, whereas a decrease was found for high dream recallers (see
Table I including t-tests for dependent samples). An analysis of variance yielded a sig-
nificant effect for the factor “recall group” (F = 301.1, p < .0001) but no substantial
difference between questionnaire and diary measure (F = 0.2, p = .8925). The interaction
between recall group and repeated measurement (Questionnaire vs. dream diary) was sig-
nificant (F = 114.8, p < .0001). In order to determine which groups contribute to the signi-
ficant interaction, three 2 × 2 ANOVAs were carried out post-hoc (similar to the ANOVA
above). Whereas the interaction for the “low recall-medium recall” pair was not signif-
icant (F = 0.5, p = .4711), the interactions for the “low recall-high recall” (F = 75.7,
p < .0001) and the “medium recall-high recall” (F = 187.3, p < .0001) pairs reached
significance.
For studying the fluctuations of each participant, the sample was also divided into three
comparable subgroups along the diary dream recall frequency: low recallers = 0 to 1 dream
per two weeks; medium recallers = 2 to 4 dreams, and high recallers = 5 to 14 dreams.
About 60.3% of the participants (N = 172) remained in their respective groups. Although
some persons changed to the adjacent group (which is partly due to the arbitrary grouping),
extreme fluctuations (changes from low to high recall or vice versa) occurred very seldom
(see Fig. 1).

Dream Content Measures

Regarding the “Emotional intensity” and “Emotional tone” variables, marked relation-
ships between the global rating scales of the questionnaire and the medians of the diary
self-ratings were found in the total sample (see Table II). With respect to the content analy-
sis data, the correlation coefficients were smaller but still significant. The global self-rating
“Creative aspects of dreams” was correlated with the mean word count of the diary dreams.
Word count was chosen since research (cf. Livingston & Levin, 1991) has shown that
P1: GCP/GYQ
Dreaming [dream] ph093-drem-366672 January 5, 2002 11:12 Style file version Nov. 19th, 1999

22 Schredl

Fig. 1. Distribution of high, medium, and low dream recallers (Questionnaire and diary measures).

creative persons tend to report longer dreams whereas other dream characteristics such as
bizarreness were not related to creativity, when dream length is statistically controlled. For
the “Realism” questionnaire item, the relationship to bizarreness/realism (content analysis)
was not significant.

Dream Content Measures for Recall Subgroups

The sample was again divided into three subgroups according to their questionnaire
dream recall frequency (cf. Table I and Fig. 1, respectively). The correlation coefficients
were considerably smaller in the low recall group that in the two other groups, with the
exception of the variable “Emotional tone” (see Table II). Although the relationship between

Table II. Correlation Coefficients Between Questionnaire (Q) and Diary Measures
Total sample Low recallers Medium recallers High recallers
Variable (N = 285) (N = 35) (N = 124) (N = 126)

Emotional intensity (Q vs. diary self-rating) .274∗∗∗ −.094 .305∗∗∗ .328∗∗∗

Emotional intensity (Q vs. content analysis) .164∗∗ −.131 .125† .299∗∗∗
Emotional tone (Q vs. diary self-rating) .348∗∗∗ .384∗ .349∗∗∗ .329∗∗∗
Emotional tone (Q vs. content analysis) .122∗ .407∗ .111 .081
Creative aspects vs. dream length .302∗∗∗ .107 .331∗∗∗ .281∗∗∗
Realism (Q) vs. Realism/bizarreness −.055 .019 −.100 −.092
(content analysis)
†p < .10. ∗p < 0.5. ∗∗p < .01. ∗∗∗p < .001; all probabilities are one-tailed.
P1: GCP/GYQ
Dreaming [dream] ph093-drem-366672 January 5, 2002 11:12 Style file version Nov. 19th, 1999

Questionnaires and Diaries as Research Instruments in Dream Research 23

Table III. Gender Differences

Women Men
Variable (N = 212) (N = 73) Effect size Statisticsa

Dream recall (Questionnaire) 4.25 ± 1.24 3.77 ± 1.43 0.36 z = 2.5 .0063
Dream recall (Dream diary) 4.17 ± 2.69 3.74 ± 2.89 0.16 t = 1.2 .1232
Dream recall (Dream diary; including 9.68 ± 3.20 8.49 ± 3.58 0.33 t = 2.7 .0042
contentless recall)
Emotional intensity of dreams (1 to 5) 3.98 ± 0.92 3.53 ± 1.09 0.44 z = 3.0 .0023
Emotional intensity (diary, self-rating) 2.68 ± 1.20 2.61 ± 1.14 0.06 z = 0.8 .4011
Emotional intensity (content analysis) 1.21 ± 1.17 1.23 ± 1.18 −0.02 z = −0.1 .9045
Emotional tone of dreams (−1 to 1) −0.04 ± 0.72 0.14 ± 0.71 −0.25 z = −1.8 .0657
Emotional tone (diary, self-rating) −0.52 ± 1.40 −0.19 ± 1.30 −0.24 z = −1.7 .0812
Emotional tone (content analysis) −0.60 ± 0.96 −0.59 ± 0.91 −0.01 z = −0.4 .7225
a t-test
(t), Mann-Whitney-U-test (z).
All probabilities are two-tailed except for dream recall frequency.

the item “Realism” and the bizarreness/realism content scale was somewhat larger for the
medium and high recall group, the correlation coefficients did not reached significance.
Again, no relationship was found for the low recall group. Using the Fisher-z-transformation,
statistical tests for the differences between the correlation coefficients of the three subgroups
were carried out. The following three comparisons reached significance ( p < .05): the
difference for the emotional intensity (questionnaire vs. diary self-rating) between low
recallers and medium recallers as well as between low recallers and high recallers. For the
emotional intensity (questionnaire vs. content analysis), only the difference between low
recallers and high recallers was significant.

Gender Differences

In Table III, gender differences concerning dream recall are depicted. Women reported
recalling their dreams more often than men (questionnaire measure), but no difference was
found for the number of dreams recorded in the dream diary. If the category “dreamt but no
explicit recall of dream content” was added to the number of morning with explicit dream
recall, the gender difference was again significant (see Table III).
Regarding gender differences in emotional intensity of dreams, women estimated their
dreams as more intense than men on the questionnaire but no difference emerged for the
diary self-ratings and the content analysis scales. For the variable “Emotional tone,” the
questionnaire item and the diary self-ratings yielded similar gender differences (cf. effect
sizes) which were marginally significant. No substantial difference, however, was found if
content analysis data were considered.

DISCUSSION

Overall, the findings indicate a clear relationship between questionnaire and diary
measures, but this relationship is mediated by dream recall frequency and not valid for all
dream characteristics investigated in the present study.
In contrast to previous findings (Cohen, 1969; Schredl, 1991), dream recall frequency
was not elevated by the diary procedure. Although the correlation between the two mea-
sures of dream recall was high, there was a different pattern depending on the initial dream
P1: GCP/GYQ
Dreaming [dream] ph093-drem-366672 January 5, 2002 11:12 Style file version Nov. 19th, 1999

24 Schredl

recall frequency: low and medium dream recallers increased their dream recall, whereas
high recallers reported less dreams in the diary as were estimated while answering the ques-
tionnaire. This drop in dream recall frequency was not previously reported in the literature.
First, this phenomenon might partly be explained by a statistical issue called tendency
to the mean. Repeated measurements of extreme values (extreme high or low) tend towards
being closer to the sample’s mean value due to general measurement error and fluctuations
within the variable. In order to estimate the contribution this effect might have, it will be
necessary to carry out retest studies with both types of dream recall measures in order to
obtain reliability coefficients. Whereas the fluctuations in dream recall frequency can be
considered as relatively small (two week period), the measurement errors of the scales may
contribute to the effect of the tendency to the mean value. In addition, the increase in dream
recall frequency in the medium recall group provides further evidence that the results could
not be solely attributed to the statistical issue.
The increase in dream recall (low and medium recallers) which was also reported in
previous studies (e.g. Cory et al., 1975) is probably an effect of focusing the attention on
dreams (cf. Schredl, 1999). The drop in dream recall present in the high dream recallers
might be explained by motivation (participants are asked to record a lot of dreams) or by
biases occurring when retrospective estimates of dream recall frequency were elicited (e.g.
adding the episodes of having dreamt but without explicit recall of dream content). To
elaborate on this topic, detailed interviews should be carried out at the end of a recall study.
In addition, it may be fruitful to investigate whether “estimation” styles (over-estimators vs.
under-estimators regarding questionnaire items) which might be connected to personality
traits related with dream recall might be an alternative explanation of the present findings
regarding changes in dream recall.
The different measurement technique (questionnaire vs. diary) clearly affects the find-
ings regarding gender differences. Whereas the questionnaire measure revealed a significant
difference and, thus, is in line with previous findings (e.g., Borbely, 1984; Giambra, Jung &
Grodsky, 1996; Schredl, 2000), the diary dream recall did not differ between the sexes. It
is difficult to decide which finding is more valid since both measures may be biased, e.g.
errors in remembering the accurate number of mornings with dream recall (cf. significant
gender difference if contentless recall was added) in response to a retrospective measure or
the influence of the diary procedure on dream recall frequency. Further research including
different measures of dream recall is needed.
Regarding the relationship between questionnaire items measuring aspects of dreams
and the diary measures, the coefficients were significant with the exception of “realism/
bizarreness,” i.e., the results were more pronounced than the data reported by Bernstein,
Belicki and Gonzales (1995). The coefficients, however, were of medium effect size which,
on the one hand, might be explained by general measurement errors of the scales and
which, on the other hand, was due to the low stability of dream content (cf. Schredl, 1998b).
If series of 20 dreams will be included into the content analysis, the magnitude of the
correlation coefficients should increase.
Regarding the considerably lower correlation coefficients between questionnaire mea-
sures and dream content scales, one must take the findings of Schredl and Doll (1998) into
consideration who have shown that dream emotions are underestimated by external judges
in comparison to the dreamer’s self-ratings, i.e., the validity of dream content analysis is
limited in this aspect. Whereas the relationships were close for dream emotions, a correlation
P1: GCP/GYQ
Dreaming [dream] ph093-drem-366672 January 5, 2002 11:12 Style file version Nov. 19th, 1999

Questionnaires and Diaries as Research Instruments in Dream Research 25

between the global estimate of dream realism and bizarreness/realism (dream content analy-
sis) was not detected. One might conclude that different characteristics of dream can be
measured with different quality, i.e., the more complex the dream characteristic is (e.g. see
question of Bernstein & Roberts (1995) cited in the introduction), the higher is the chance
that biases affect the person’s global estimate. For further elaboration, future studies should
investigate a large variety of dream aspects (e.g. Gruber, 1988).
If one examine the effect of the measurement technique on gender differences in dream
content, a similar pattern was found for intensity of dream emotions as reported above for
dream recall frequency. Whereas women reported more intense dream on the questionnaire,
diary data showed no differences. Regarding the emotional tone, the magnitude of the gender
difference was comparable. With respect to the content analysis data, the low validity of the
measurement of dream emotions may also have accounted for the lack of substantial gender
differences. The same question as stated above arises: which finding should be considered
as more valid?
Another important methodological problem should be addressed in future studies: the
relationship between questionnaire and diary data is depending from the level of dream
recall. Aside the finding regarding emotional tone of dreams, the correlation coefficients
of the low recall group were considerably smaller than those of the two other groups. This
could not fully explained by the lower measurement quality (smaller number of dreams per
low dream recaller) since emotional tone (questionnaire measure) was strongly related to
the emotional tone of the diary dreams. These findings indicate rather that recall processes
which the person must carry out to respond to a global questionnaire item measuring dream
content are of importance. i.e., the assumption of Gruber (1988) that biases can occur in
low dream recallers in respect to dream questionnaire data, is supported by the findings
of the present study. In order to provide data demonstrating sufficient validity of dream
questionnaire items, further research is needed.
To summarize, the relationship between questionnaire and diary measures is strong for
at least some dream characteristics such as dream emotions if one take the possible reduction
of explained variance by general measurement errors and variability of dream content into
account. The relationships, however, are much weaker for low dream recallers indicating
that recall and sampling processes carried out by the participants while answering the item
should be studied in a more detailed way. Using the example of testing gender differences,
the findings of the present study clearly indicate that the measurement technique affects the
results. Future research is necessary to prove the validity of dream questionnaires.

REFERENCES

Agargün, M. Y., Kara, H., Bilici, M., Cilli, A. S., Telci, M., Semiz, Ü. B., & Basoglu, C. (1999). The Van Dream
Anxiety Scale: a subjective measure of dream anxiety in nightmare sufferers. Sleep and Hypnosis, 1, 204–211.
Baekeland, F. (1970). Correlates of home dream recall: I.REM sleep in the laboratory as a predictor of home dream
recall. Journal of Nervous and Mental Disease, 150, 209–214.
Belcher, M., Bone, R. N., & Montgomery, D. D. (1972). Rigidity and dream recall. Psychological Report, 30, 858.
Bernstein, D. M., & Belicki, K. (1995–96). On the psychometric properties of retrospective dream content ques-
tionnaires. Imagination, Cognition and Personality, 15, 351–364.
Bernstein, D. M., Belicki, K., & Gonzales, D. (1995). The development and assessment of the reliability and
validity of a dream content questionnaire (DCQ). Sleep Research, 25, 138.
Bernstein, D. M., & Roberts, B. (1995). Assessing dreams through self-report questionnaires: relations with past
research and personality. Dreaming, 5, 13–27.
P1: GCP/GYQ
Dreaming [dream] ph093-drem-366672 January 5, 2002 11:12 Style file version Nov. 19th, 1999

26 Schredl

Borbely, A. (1984). Schlafgewohnheiten, Schlafqualität und Schlafmittelkonsum der Schweizer Bevölkerung:

Ergebnisse einer Repräsentativumfrage. Schweizerische Ärztezeitung, 65, 1606–1613.
Bruni, O., Lo Reto, F., Recine, A., Ottaviano, S., & Guidetti, V. (1999). Development and validation of a dream
content questionnaire for school age children. Sleep and Hypnosis, 1, 41–46.
Cohen, D. B. (1969). Frequency of dream recall estimated by three methods and related to defense preference and
anxiety. Journal of Consulting and Clinical Psychology, 33, 661–667.
Cohen, D. B., & MacNeilage, P. F. (1974). A test of the salience hypothesis of dream recall. Journal of Consulting
and Clinical Psychology, 42, 699–703.
Cory, T. L., Orniston, D. W., Simmel, E., & Dainoff, M. (1975). Predicting the frequency of dream recall. Journal
of Abnormal Psychology, 84, 261–266.
Domino, G. (1982). Attitudes toward dreams, sex differences and creativity. Journal of Creative Behavior, 16,
112–122.
Early, E. M. (1977). Dream recall: Frequency and variation. Dissertation Abstracts International, 38, 2857-B.
Giambra, L. M., Jung, R. E., & Grodsky, A. (1996). Age changes in dream recall in adulthood. Dreaming, 6,
17–31.
Goodenough, D. R., Shapiro, A., Holden, M., & Steinschriber, L. (1959). A comparison of “dreamers” and
“nondreamers.” Journal of Abnormal and Social Psychology, 59, 295–302.
Gruber, R. E. (1988). Dreaming style and waking personality. University of Cincinnati: Unpublished dissertation.
Kallmeyer, R. J., & Chang, E. C. (1997). The multidimensional dream inventory: preliminary evidence for validity
and reliability. Perceptual and Motor Skills, 85, 803–808.
Kramer, M., & Roth, T. (1979). The stability and variability of dreaming. Sleep, 1, 319–325.
Kroth, J., Gilbert, H., Guichard, A., & Quatman, T. (1999). Analysis of factor structure in a dream inventory.
Perceptual and Motor Skills, 89, 657–658.
Kroth, J., Yoneda, D., & Hammond, A. (1998). Fear of fat and the dream process: a correlational investigation.
Psychological Reports, 83, 1197–1198.
Livingston, G., & Levin, R. (1991). The effects of dream length on the relationship between primary process in
dreams and creativity. Dreaming, 1, 301–309.
Moffitt, A., Hoffmann, R., Wells, R., Armitage, R., Pirgeau, R., & Shearer, J. (1982). Individual differences among
pre and post awakening EEG correlates of dream reports following arousals from different stages of sleep.
Psychiatric Journal of the University of Ottawa, 7, 111–125.
Reed, H. (1973). Learning to remember dreams. Journal of Humanistic Psychology, 13(3), 33–48.
Schredl, M. (1991). Traumerinnerungshäufigkeit und Trauminhalt bei Schlafgestörten, psychiatrischen Patienten
und Gesunden. Universität Mannheim: unveröffentlichte Diplomarbeit.
Schredl, M. (1998a). Träume und Schlafstörungen: Empirische Studie zur Traumerinnerungshäufigkeit und zum
Trauminhalt schlafgestörter PatientInnen. Marburg: Tectum.
Schredl, M. (1998b). The stability and variability of dream content. Perceptual and Motor Skills, 86, 733–734.
Schredl, M. (1999). Dream recall: research, clinical implications and future directions. Sleep and Hypnosis, 1,
72–81, A2–A4.
Schredl, M. (2000). Gender differences in dream recall. Journal of Mental Imagery, 24, 169–176.
Schredl, M., & Doll, E. (1998). Emotions in diary dreams. Consciousness and Cognition, 7, 634–646.
Schredl, M., Schäfer, G., Hofmann, F., & Jacob, S. (1999). Dream content and personality: thick vs. thin boundaries.
Dreaming, 9, 257–263.
Schredl, M., Schröder, A., & Löw, H. (1996). Traumerleben von älteren Menschen—Teil 2: Empirische Studie
und Diskussion. Zeitschrift für Gerontopsychologie und -psychiatrie, 9, 43–53.
Soper, B., Ghramani, P., Gombossy, T., & Dowling, J. (1990). A preliminary factor analysis of dream themes.
Louisiana Journal of Educational Research, 15, 9–17.
Zadra, A. L., & Nielsen, T. A. (1999). The 55 typical dream questionnaire: consistency across three student
samples. Sleep Supplement, 22, S175.