Methods

One hundred and sixty~ne female college students completed three questionnaires as part of a larger study, the Hoarding Scale, the Magazine Questionnaire, and the Newspaper Questionnaire.

The Hoarding Scale is a 24 item version of the original Frost and Gross (1993) Hoarding Scale. This measure asks subjects to indicate the extent to which they agree or disagree with each of the items from 1 (strongly disagree) to 5 (strongly agree). Examples of items include, *I have trouble throwing things away* and *I am worried I may throw things away that later I will find valuable,* and *I have saved so much stuff that it is too much to manage.* This scale has been found to he a reliable and valid measure of hoarding behaviors in student, community, and clinical samples.

The Magazine Questionnaire (MQ) and Newspaper Questionnaire (NQ) were designed to measure thoughts that occur in a hoarding related context. The instructions ask subjects to imagine that they have purchased a magazine (or newspaper), kept it for a month (or week), and have read some, but not all, the articles it contains. Thken the instructions ask subjects to consider whether to keep or discard the magazine (newspaper). Subjects are instructed to indicate the extent to which they are likely to have each thought listed on the questionnaire when making this decision. Nine of the thoughts are reasons for saving the target possession, and nine thoughts are reasons for discarding the target possession. In addition, subjects are asked to indicate the value they are likely to assign to this item, and to give a judgement of the likelihood they would discard it. See table 1 for sample items.

Results

Reliability analyses indicated that the Reason to Save and Reason to Discard measures were reliable for hoth the MQ and NQ. The alphas ranged from .85 to .89 for the 4 measures. Correlations between the four measures on the MQ and NQ ranged from .39 to .59. Because of these moderate correlations, analyses were done separately for the MQ and NQ.

Correlational analyses, using p < .01 as the level of significance, indicated that the Magazine Questionnaire measures were more closely related to hoarding phenomena than the Newspaper Questionnaire. For both the MQ and NQ, the frequency of reasons to save thoughts were significantly and positively correlated with the Hoarding Scale (see Table 2). For the MQ (but not the NQ) the Hoarding Scale was negatively correlated with the frequency of Reasons to Discard, positively correlated with Value given to the magazine, and negatively correlated with the Intention to Discard the magazine.

Frequency of Reason to Save thoughts was negatively correlated with the frequency of Reason to Discard thoughts for MQ, but not for NQ (See tables 3 and 4). Frequency of Reason to Save thoughts was negatively correlated with Intention to Discard for both the MQ and NQ. Frequency of Reason to Discard thoughts was positively correlated with Intention to Discard for both MQ and NQ.

Stepwise multiple regression analyses were conducted predicting the Intention to Discard by frequency of Reason to Save thoughts, the frequency of Reason to Discard thoughts, and the Hoarding Scale for both MQ and NQ. Tables 5 shows the results of these regressions. For both analyses, both the frequency of Reason to Save thoughts and Reason to Discard thoughts contributed significandy and independendy to the equation. The equation accounted for 64% of the variance in Intention to Discard the magazine and 53% of the variance in Intention to Discard the newspaper. The Hoarding Scale did not enter either equation.

Discussion

All hypotheses were supported by the finding from the Magazine Questionnaire. Partial support for the hypothesis was obtained from the Newspaper Questionnaire. The findings suggest that this imagery-based task can be used to study hoarding related phenomena, especially the Magazine Questionnaire. Similar imagery-based tasks may provide a method for testing hypotheses derived from the cognitive behavioral model of compulsive hoarding (Frost & Hartl, 1996). Similar data from clinical populations are needed to flirer validate this task.

These findings suggest that reasons to save a possession are more prominent in the thinking of people who display hoarding behavior than those who do not, Furthermore, although people with high scores on the hoarding Scale thought about fewer reasons to discard the object than people scoring low on this scale, the differences were not as large as they were for reasons to save. While more frequent reason to save thoughts characterized people who hoard to a gater extent than less frequent reasons to discard, both measures contributed significandy and independendy to the prediction of the Intention to Discard a possession. Implications for a cognitive-behavioral treatment of hoarding are that in a program that focuses on modification of reasons to save and discard possessions, more emphasis should be placed on reducing reason-to-save thoughts than reason-to discard thoughts.

References

Frost, R. & Gross, R. The hoarding of possessions. Behaviour Research and Therapy, 31 367-381, 1993.

Frost, R. & Hartl, T. A cognitive-behavioral model of compulsive hoarding. Behavior Research and Therapy, 34 341-350, 1996.

Frost, R., Krause, M. & Steketee, G. Hoarding and obsessive compulsive symptoms. Behavior Modification,20, 116-132, 1996.


Table 1. Magazine Questionnaire (abridged)

Imagine that you have purchased a magazine, kept it for a month, and read some, but not all, of the articles. You are now trying to decide whether or not to throw the magazine away, even though there are things in it you would like to keep. To what extent would you be likely to have each of the following thoughts when making your decision (1 = not at all; 7 = very much)?

1. I may lose something important by throwing it away.

1 2 3 4 5 6 7

2. To keep this would require too much room.

1 2 3 4 5 6 7

3. If I throw it away, I might not remember what was there.

1 2 3 4 5 6 7

4. I can always find something else to substitute.

1 2 3 4 5 6 7

5. What level of value would you place on this magazine?

0...........1............2............3............4...........5........... 6..............7

No Value...............................Moderate.........................Very High Value

6. How likely would you be to throw this magazine away?

0.............1.............2.............3.............4............. 5.............6.............7

Very unlikely-..................................................................... ....Very likely-

I would definitely keep it.............................I would definitely throw it away

Table 2. Correlations between the hoarding scale and Reasons to Save, Reasons to Discard, Value, and Intention to Discard for the Magazine and Newspaper tasks.

Table 3. Intercorrelations of Magazine Questionnaire measures

Table 4. Intercorrelations of Newspaper Questionnaire measures

Table 5. Multiple regressions predicting the Intention to Discard from Reasons to Save, Reasons to Discard, and the Hoarding Scale.