Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)




Although many authors have used the term "grief work" since Lindemann (1944) first coined it, there has been no systematic investigation of the nature of this phenomenon. This study sought to delineate the cognitive and interpersonal events that occur during grief events and grief resolution. It was hypothesized that grief work involves the creation of a new relationship with the deceased, and that this would be manifested in a variety of styles of grief.

Men and women who had been widowed within the previous two years were identified through newspaper obituaries and recruited as subjects. Sixty subjects completed two process measures. The Grief Work Survey measured recent behaviors and thoughts concerning the loss. The Attitude Toward Grief Survey measured attitudes toward loss and grief. Subjects also completed forms designed to measure a number of prebereavement variables and to assess the degree of grief resolution. In addition an unstructured interview was conducted.

On the Grief Work Survey, subjects reported that they thought about their loss quite frequently. However, the loss was only rarely mentioned by other people and subjects stated that they only rarely discussed the loss with others. This pattern of behavior contrasted sharply with subjects' attitudes toward grief. Subjects reported on the Attitudes Toward Grief Survey that under optimal conditions grief is shared with other people.

In order to determine styles of grieving, a hierarchical clustering of subjects was conducted based on responses to the Grief Work Survey. When clusters were based on the entire form, four clusters emerged. These were labeled "nongrievers," "overwhelmed grievers," "stiff upper lip grievers" and "emotionally expressive grievers." The overwhelmed and stiff upper lip clusters stood out as showing significantly poorer outcomes. There were few differences among groups on prebereavement variables.

In a second attempt to determine styles of grief responses to the Grief Work Survey were factor analyzed. Four factors emerged. These appeared to lie along two dimensions: public-private and pain-comfort. Factor scores for each subject on each factor were calculated and these were correlated with the various pre-bereavement and outcome measures. The pain factors were found to be correlated with bad outcome. A long terminal illness was correlated with a private pattern of grief work. Finally, older women who reported high marital satisfaction were found to engage most frequently in behaviors which loaded highly on the comfort factors.

Behavior during the unstructured interview was reported. Implications concerning a cognitive theory of grief and concerning appropriate therapeutic interventions for the bereaved were discussed.