Dissertation on Grief when Losing Someone to Suicide

Suicide bereavement: a unique bereavement experience?

Abstract

Whether a suicide survivor’s grief is different from the grief experienced by survivors of accidental and natural deaths has been a popular research topic in the past. The results of these research studies have been conflicting with some finding there is a significant difference (Bailley, Kral & Dunham, 1999) and others failing to find this difference (Cleiren, Diekstra, Kerkhof & Van der Wal, 1994). However, research investigating individual’s perceptions of these different survivors is lacking. Therefore, the aim of this study was to investigate whether participant’s perceptions of a survivor’s grief reactions differed depending upon the mode of death they were exposed to. Participants were 252 university students who read a hypothetical scenario regarding a death within a family and were randomly assigned to one of five possible deaths; suicide expected, suicide unexpected, accidental, natural expected and natural unexpected. The Grief Experience Questionnaire (GEQ) was used to measure participant’s perceived grief reactions. Results of the current study found that participants who were exposed to either of the suicide deaths reported higher scores on the GEQ than those exposed to the other modes of death. These results suggest that suicide survivors are perceived to have more intense and a higher frequency of unique grief reactions than survivors of other modes of death. This research provides an implication of how the public might support someone they know who has been bereaved by suicide and highlights the need for an overall better understanding of the grief reactions specifically experienced by suicide survivors.

Introduction

An act of suicide is an intentional cause of one’s own death. The total number of deaths registered in the UK for 2015 was 602,782 (Office for National Statistics, 2016a) and the number of these resulting from suicide was 6,188 (Office for National Statistics, 2016b). Both numbers have risen significantly since 2011 when the total number of deaths registered was 552,232 and the total number of suicides was 6045. The suicide rates of women living in the UK are currently the highest they have been in a decade (Samaritans, 2017) therefore the present study will include a hypothetical scenario involving a female death.

Grief

Grief is an individual’s response to bereavement and includes various psychological and physical symptoms that change over time (Solomon & Shear, 2015). Grief is unique to the individual and results in a multitude of complex responses (Cowles & Rodgers, 1991;

Cleiren, 1993). When a loved one has died, most bereaved persons experience intense grief, but the majority adapt well over time and some even show few signs of distress (Bonanno, Wortman & Nesse, 2004). However, for some people, it is harder to reach an acceptance of the death and a minority suffer from a much more intense and distressing form of grief known as ‘complicated grief’.

Complicated Grief

When an individual’s grief becomes complicated it can have a severe impact on their ability to function and disrupts their overall quality of life. This form of grief includes severe separation distress, intense longing for the person who has died (Prigerson et al, 2009), preoccupation with the deceased and difficulty accepting the loss (Boelen & van den Bout, 2005). Research has suggested that complicated grief reactions are likely to occur with higher frequency for those who have lost someone to suicide (Dyregrov, Nordanger & Dyregrov, 2003) and therefore these individuals are at risk of various negative health outcomes, such as suicidal ideation, that have been found to be associated with complicated grief (Latham & Prigerson, 2004).

Risk Factors

To understand why some individuals suffer from more frequent and intense grief reactions, research has focused on the potential risk factors that may cause this to occur. Stroebe and Schut (2001) suggested that situational, personal and interpersonal factors impact how an individual can cope with grief. Further research has proposed an integrative risk factor model that includes these factors (Stroebe, Folkman, Hansson & Schut, 2006). The aim of developing this model was to be able to identify bereaved individuals who are more at risk of suffering from a higher frequency of intense grief reactions. This model has suggested that factors such as the mode of death, relationship to the deceased, social support and religiosity can impact how well an individual is able to cope with their grief. Understanding the risk factors that can prolong or intensify an individual’s grief is important in being able to identify those who would need and benefit from professional help (Schut, Stroebe, van den Bout & Terheggen, 2001).

Relationship to Deceased

As previously mentioned, the closeness of the relationship an individual had with the deceased is a factor that impacts their grief reactions. Research has suggested that the nature of an individual’s grief is dependent upon the bonds that were severed and the importance of these bonds (Weiss, 1999). Losing a primary attachment figure (child, spouse or parent) has been linked to a more prolonged process of grief and more intense separation distress than losing an extended family member or friend (Holland & Neimeyer, 2011). The loss of these first-degree relatives has also been found to increase the amount of complicated grief symptoms and severe grief reactions for bereaved individuals, and can cause difficulty in making meaning out of the loss when compared to extended relatives (Holland, Currier & Neimeyer, 2014; Prigerson et al., 2002; Servaty-Seib & Pistole, 2007).

Spousal bereavement involves the loss of an important attachment and can have life-changing impacts on an individual. Marris (1958) suggested that one of the reasons why losing a spouse has such an extreme impact on an individual’s ability to function is because this loss can deprive their life of meaning. The loss of a spouse has been described as one of the most stressful life events an individual can face (Prigerson et al., 1995). This loss means that an individual loses someone they are dependent upon whether this is for love, support or financial stability and these impacts can have negative effects on how an individual grieves. Research comparing bereaved spouses with bereaved parents and a non-bereaved group found that bereaved spouses, particularly widows, reported lower levels of overall life satisfaction than those in a non-bereaved comparison group (Arbuckle & de Vries, 1995). This finding emphasizes the damaging effects spousal bereavement has on an individual’s life.

Mode of Death

Unexpected and sudden deaths can facilitate an unusually long and intense denial of the loss that can delay the grieving process (Wadland, Keller, Jones & Chapados, 1988). Unanticipated deaths prevent any preparation for life after the death, and can potentially leave individuals with uncertainty of where to turn. Suddenness of the death often means loved ones lose the opportunity to say goodbye (Kristensen, Waisæth & Heir, 2012) possibly leaving them with unresolved conflicts that complicate their grief. Having little or no forewarning of a death can intensify grief reactions and lead to overall poorer functioning for the bereaved (Parkes & Weiss, 1983; Rando, 1993).

Sudden deaths can sometimes be violent and can take the form of suicides, accidents or homicides (Farberow, Gallagher-Thompson, Gilewski & Thompson, 1992).  Research has found that those bereaved by suicide present grief reactions that differ from those bereaved by other modes of death (Demi, 1984; Ellenbogen & Gratton, 2001; Jordan, 2001). However, the reactions following a suicide such as shock, guilt and shame have also been found to follow after an accidental death (Séguin, Keily & Lesage, 1993). It could be possible that the reactions caused by suicide are only severe in the short term as McIntosh (1993) suggested that the survivors of suicide are similar to those of sudden deaths and any difference in grief reactions shown by suicide survivors become negligible after the second year. This suggests that whilst the grief of suicide is more complicated at first, the impact of a sudden loss, regardless of how it occurred, is the reasoning behind prolonged and severe grief reactions.

Unique Grief Reactions

Nevertheless, survivors of suicide are likely to have grief reactions unique to those of other deaths. Feelings of rejection, stigma and blame are more likely to be present in those that have lost someone to suicide and may even result in them hiding the cause of death from others (Sveen & Walby, 2008). These researchers performed a systematic review comparing suicide survivors’ reactions with the reactions of survivors of other modes of death and found that these unique reactions were more frequently reported by the survivors of a suicide. The results from this study support previous findings that suicide bereavement is distinct from any other forms of bereavement (Jordan, 2001).

Measuring Grief Reactions

In order to measure the differences in grief reactions between suicide survivors and other suicide groups, the Grief Experience Questionnaire (GEQ) was developed (Barrett & Scott, 1989). The development of this questionnaire was necessary as previous grief reaction measures such as the Grief Experience Inventory (Sanders, 1983), have been criticised for being too general or too narrow to use in the study of suicide specific reactions. The GEQ measures the grief reactions expected in any bereavement and those unique to suicide survivors. Barrett and Scott’s GEQ included 11 subscales; somatic reactions, general grief reactions, search for explanation, loss of social support, stigmatization, guilt, responsibility, shame, rejection, self-destructive behaviour and unique reactions and consisted of 55 items.

The strengths of the GEQ have made it an increasingly popular tool when comparing the grief reactions in suicide survivors to those in survivors of other modes of death. The first of these strengths is the high internal consistency reliability of each subscale and the total GEQ scale that ranged from moderately high to high suggesting that it is well designed to measure all groups of survivors. Other strengths of the questionnaire include its relatively quick completion time and its demonstrated capability of successfully differentiating between survivors.

Suicide

Survivors of suicide are those who have lost someone to a successful suicide. Survivors not only include family and friends, but also co-workers, classmates and therapists (McIntosh, 1987a). Both clinicians and researchers have suggested that the mourning process following a suicide is unique and more difficult than the mourning process following other modes of death (Clark & Goldney, 1995; Sprang & McNeil, 1995). This suggests that the grief following a suicide requires special attention and that the therapy and counselling provided to survivors of other deaths may not be helpful to suicide survivors. Losing a loved one to suicide has been highlighted as a risk factor for developing complicated grief (Rando, 1993) as suicide survivors struggle more with meaning making of the death (Grad & Zavasnik, 1996). If a suicide is unexpected, survivors may not know the reason for it and may therefore struggle with understanding why it happened.

One study comparing suicide survivors with survivors of other deaths; accident, unanticipated natural and anticipated natural, found that those bereaved by suicide experienced more feelings of rejection and responsibility, and that they showed more unique and total grief reactions (Bailley et al., 1999). Participants within this study were 350 Canadian university students who had previously been bereaved by one of the four modes of death. The Grief Experience Questionnaire (GEQ; Barrett & Scott, 1989) was used to measure the different dimensions of participant’s grief. Although this study used a relatively large sample and used a comparison group, the relationship to the deceased was not controlled as participants were individuals who had lost either non-immediate family members or first degree family members. This is a potential limitation of the study as the closeness of the relationship shared with the deceased may have also had an impact on the participants grief and therefore it cannot be assumed that it was solely the mode of death that caused any differences.

Whilst the results from this study are similar to those of previous research (Kovarksky, 1989; Reed & Greenwald, 1991), there is also evidence to suggest that a suicide survivors process of grief is no more complicated or prolonged than that of other survivor groups (Cleiren et al., 1994; Van der Wal, 1990;). More contemporary research has found that a significant difference between survivors of suicide and survivors of natural deaths is non-existent (Callahan, 2000). When comparing 210 participants who had recently been bereaved, results collected with the use of the GEQ found that losing the deceased to suicide had an insignificant effect on an overall level of grief. Nevertheless, the sample used within this study consisted of bereaved individuals who attended support groups and therefore had already sought out help. Consequently, the results for this study cannot be generalised to all bereaved individuals as though who have not attempted to seek help may be suffering more than those that have. This contrasting evidence suggests that further research is necessary to understand the differences in grief reactions of survivors for different modes of death.

Stigmatization

Worden (1991) suggests that there is a stigma associated with suicide in our society. Committing suicide has been reported to be stigmatized as cowardly, irresponsible and narcissistic (Maris, 1981) and because of this, suicide survivors are likely to also feel stigmatized. Public perception research has found that participants were likely to hold less favourable perceptions of suicide survivors and saw them as less likeable and more accountable for the death (Allen, Calhoun, Cann & Tedeschi, 1994; Calhoun, Selby & Faulstich, 1980). When reviewing suicide and the social processes that follow Jordan (2001) concluded that suicide survivors are likely to be more negatively viewed by people in their social network than survivors of other modes of death. This was found to be true when comparing suicide survivors with natural death survivors as those bereaved by suicide received significantly less emotional support from family and friends and confided in them less (Farberow et al., 1992).

Despite this, stigmatization responses for suicide survivors were found to be similar with those of traumatic death survivors (Feigelman, Gorman & Jordan, 2009). Participants within this study were parents who had lost a child to suicide, natural causes or in another traumatic way. The traumatic deaths within this sample were either caused by accidents or homicides and results showed that the mean stigmatization score of this group was not drastically different to that of the suicide group. The stigmatization felt by the natural death survivors, nevertheless, was significantly lower than the other groups. This study supports the notion that suicide survivors do feel as though they are stigmatized, but also highlights that they are not the only survivor group to feel this way. As this study used only parents as survivors, the results may not be applicable to others such as spouses and siblings as the relationship to the deceased may have had a significant impact. Stigmatization can lead to suicide survivors feeling as though they have lost their social support. Whilst there is research to support this (Fine, 1997), there is also research suggesting that support may be offered but not to the extent survivors want or they may perceive it as missing (Seguin, Lesage & Kiely, 1995; Van Dongen, 1993). This can lead to “self-stigmatization” (Dunn & Morrish-Vidners, 1988) were suicide survivors assume that people are judging them negatively, resulting in them withdrawing from their social groups.

Guilt

Guilt is another grief reaction presented by suicide survivors as they often feel as though they should have or could have done something to prevent the death from happening (Worden, 2010). Guilt has been found to be present in most survivor groups (Parkes, 1970) but is suggested to last longer and be more intense in those who have been bereaved by suicide (Battle, 1984; McIntosh & Kelly, 1992). Survivors of suicide may also feel guilty if they think they were a reason as to why their loved one committed suicide. These feelings of guilt are problematic and can lead to complications in the grieving process for survivors as they may ruminate more over what they could have done to prevent the death. Although guilt is stated to be stronger and more persistent in suicide survivors, there is evidence to suggest that this reaction is no more heightened for these individuals than it is for survivors of other deaths (Barrett & Scott, 1990).

Responsibility

These feelings of guilt suggest there is also a sense of responsibility for the death. In Bailey et al. (1999) study the results showed that suicide survivors reported more feelings of responsibility for the death when compared with survivors of both natural deaths and the accidental death. This finding is supported by similar results (Miles & Demi, 1992; Silverman, Range & Overholser, 1995). It has been suggested that these feelings develop if suicide survivors feel responsible for not recognising the signs that suicide was inevitable or for not speaking about them if they did (Bernhardt & Praeger, 1983). This feeling of responsibility places a burden on survivors that can lead to difficulties in trying to understanding the reasoning for the suicide and can result in a more prolonged grief process.

Shame

With so much research suggesting that suicide is heavily stigmatized, it has also been concluded that suicide is perceived as a shameful event (Ginsburg, 1971). Feeling ashamed or embarrassed about the death may lead to survivors hiding the cause or lying about it being a suicide. Research has found this to be true with up to 44% of suicide survivors lying to others about how a loved one has died (Range & Calhoun, 1990).  Modern research has found that suicide survivors report more feelings of shame and a need for concealing the cause of death (Sveen & Walby, 2008) Denial over the cause of death can result in survivors feeling uncomfortable speaking to their peers or professionals (Calhoun, Selby & Selby, 1982) leaving them to deal with their grief alone and could possibly explain why self-stigmatization occurs.

Rejection

Rejection following a death refers to the feeling of being rejected or abandoned by the deceased. The death of a loved one by any mode can cause feelings of rejection for the survivors but, those who are bereaved by natural or accidental deaths are more likely to eventually understand that the deceased did not leave them intentionally (Schuyler, 1973). This rational understanding is not as easy to overcome for suicide survivors as the very act of suicide is an act of rejection by the deceased of their life and those around them. It is therefore plausible that suicide survivors feelings of rejection are more intense and severe (Yufit, 1977) and universal than those bereaved by other causes of death (Osterweis, Solomon & Green, 1984).

Search for Explanation

If the reason behind the suicide is unknown, acceptance of the death may not be so easily reached by those bereaved. Accepting the death is an important process in helping survivors recover for the loss. However, searching for an explanation as to why a suicide occurred is more intense and less easily resolved for these survivors than it is for survivors of other deaths (Wallace, 1977). Nevertheless, this statement has been refuted by research comparing suicide survivors with natural death survivors showing that the groups did not differ when measured with the GEQ subscale ‘search for explanation’ (Harwood, Hawton, Hope & Jacoby, 2002).

Methodological Limitations

Research studies examining the differences in grief reactions between suicide survivors and other death survivors have been criticised for having significant methodological weaknesses (McIntosh, 1987b). One of these weaknesses is that the samples of participants have been relatively small, for example in Barrett and Scott’s (1990) study the total number of participants was 57. The use of a small sample means the results would yield little power. A reason as to why the research so often uses small samples is due to the fact that participation is voluntary and, due to the sensitive nature of the topic, the amount of volunteers may be low.

A further limitation is that participants within a sample are often from clinical settings or support groups and have already sought help with their grief. This means that the results from these studies cannot be generalised to those who haven’t volunteered and possibly refused to participate. This sample of people could potentially be the largest group of survivors (Van Dongen, 1990) and so research could be failing to investigate the majority of bereaved individuals. In order to increase sample sizes, researchers have begun to use participants who have not been bereaved and asked them to provide their perception of survivor’s grief reactions. Perception studies provide information on how the public would view certain deaths and the survivors but researchers have to be careful not to generalise these results to actual grief reactions.

Rationale

Public perceptions of suicide survivors are important as these individuals could potentially become part of a suicide survivors support group, and so results from this research could give an indication of how they may provide this support. Further research comparing the perceptions of grief of different modes of death is necessary as studies in this area is lacking. The need for more recent research in this area provides the present study with its rationale. Another reason for the partaking of the current research is to examine whether the participant’s perceptions of suicide survivors change when the suicide was expected compared to when it was unexpected.

Hypothesis

After consideration of the previous research regarding the differences in grief between survivors of various deaths it is hypothesised that participants will perceive suicidal death as a significantly unique grieving experience and those presented with either of the suicide vignettes will report higher GEQ scores.

Method

Design

The design of this study is experimental using independent measures as each participant was shown 1 of 5 hypothetical scenarios. The independent variable is the 5 modes of death; expected natural, unexpected natural, accidental, expected suicide and unexpected suicide. The dependent variable is the participant’s perception of grief which is measured by the GEQ (Barrett & Scott, 1989).

Participants

The total number of participants recruited was 252. The participants were recruited through opportunity sampling via the SONA system that was available to first year psychology students. Students completed the study to achieve credits for their Research Methods module.

Materials/apparatus

The questionnaire that participants answered was created on Qualtrics.com. The vignette provided to participants was a hypothetical story about the lives of Michael and Laura and their 3 children. Information is provided to the participants about the couple’s history, their jobs, their children’s ages and their pets (see Appendix A). The vignette ends with Michael entering the kitchen to find Laura dead on the floor. The participants will read one of five hypothetical deaths. The suicide expected death involved Laura taking an intentional overdose after suffering with depression for the past 12 years, the suicide unexpected death was again an intentional overdose but included no further information, the accidental death involved Laura falling and hitting her head on the kitchen worktop, the natural expected death involved Laura dying from breast cancer which she had previously been diagnosed with and the natural unexpected death involved Laura dying from a sudden brain haemorrhage.

The GEQ (Barrett & Scott, 1989) measures 11 dimensions of grief consisting of 55 items. This version of the GEQ has been modified to 8 factors after a factor analysis found these to be more meaningful (Bailley et al., 2000) resulting in the removal of the three subscales ‘general grief reactions’, ‘loss of social support’ and ‘unique reactions’. The 8 factors proposed by the authors were ‘abandonment/rejection’, ‘stigmatization’, ‘search for explanation’, ‘guilt’, ‘somatic reactions’, ‘responsibility’, ‘self-destructive orientation’ and ‘shame/embarrassment’.

The GEQ used within the questionnaire (see Appendix B) in the current study alters this modification slightly to consist of only 6 factors with ‘somatic reactions’ and ‘self-destructive orientation’ being removed. The total number of items used within the questionnaire has therefore decreased from 55 to 44. The wording of the statements was also adapted to apply to perceptions of grief rather than actual grief. An example of one of the statements used in the GEQ is ‘Feel as though the death was somehow a deliberate abandonment of him’ which participants responded on a Likert scale scored 1 = “never”, 2 = “rarely”, 3 = “sometimes”, 4 = “often”, and 5 = “almost always”.

Procedure

A Research and Ethics proforma was completed (see Appendix C) and approval for the study was given by Sheffield Hallam University. Once participants had chosen the current study from a list of active research on the SONA system they were presented with an information sheet (see Appendix D). Participants were told that they would be required to read a short hypothetical scenario and then complete a questionnaire that would take no longer than 15 minutes. As the study was online, participants were informed that their full completion of the study would be taken as consent to use their data. It was emphasised that participation was voluntary and that they therefore did not have to participate. Participants were informed that they could opt out if they chose to but that once they had clicked ‘submit’ they would no longer be able to withdraw their answers due to them remaining anonymous. Participants were also made aware that if they had been bereaved in the last 2 years they should not take part due to the sensitive nature of the topic.

After reading the information sheet, participants were directed to the hypothetical scenario and were randomly assigned to one of the five possible endings regarding the death of the character Laura. Upon clicking the next button, after reading the scenario, participants were presented with the adapted 44-item GEQ and were instructed to take their time and to try and answer all questions honestly. Before submitting the data, participants read through a debrief sheet (see Appendix E) thanking them for their participation and revealing to them that they had read one of five death scenarios. Contact information was also provided for both myself and my research supervisor, advice for wanting any help regarding a bereavement was also given. To complete the study participants were asked to click the ‘submit’ button.

Ethical Considerations

The current study was approved to BPS standards. Protection of the participants was maintained through the use of exclusion criteria preventing anyone who had been bereaved in the last two years from participating. Information regarding where they could seek any necessary help regarding grief was also provided. Informed consent was obtained once participants had read the information sheet and clicked ‘next’ and participants were made aware of this. The study also maintained participant’s confidentiality as, with it being an online study, their data remained anonymous and they weren’t required to provide any personal information. The results were then stored on a password protected computer that only I had access to. Before completion, participants were also informed that once they had clicked the ‘submit’ button they would no longer be able to withdraw their information and were consenting for their results to be used for analysis. Finally, after clicking ‘submit’ participants were presented with a debrief sheet providing them with details regarding the true aim of the study. The only ethical breach within the study was the use of deception as participants were not aware they would be reading one of five possible hypothetical scenarios. However, this breach was only minor and was necessary to obtain reliable results.

Results

The data collected from Qualtrics was input into SPSS version 24, this included 252 participants. The 6 subscales were presented by various numbers of statements; ‘abandonment/rejection’ – 11 items, ‘stigmatization’ – 10 items, ‘search for explanation’ – 7 items, ‘guilt’ – 6 items, ‘responsibility’ – 5 items and ‘shame/embarrassment’ – 6 items. Participant’s scores from each statement within each of the six subscales were added together to create a total score. A total score column was created to represent each subscale.

The total scores and standard deviations for the total GEQ score and for all 6 subscales were calculated and are summarised within table 1 below. The highest total score of the Total GEQ score was for the suicide unexpected mode of death (total=159.38) suggesting that participants mostly answered ‘often’ or ‘almost always’ when presented with this vignette. The lowest total score was for the natural expected mode of death (total=129.77) suggesting that participants mostly responded with ‘never’ or ‘rarely’ when presented with this vignette. The highest standard deviation score was again for suicide unexpected (SD=24.10) and the lowest was for natural unexpected (SD=19.35).

The highest totals for the six subscales were consistently from one of the suicide mode of deaths. Suicide expected had the highest total score for ‘stigmatization’ (total=33.22) with suicide unexpected having the highest total scores for the remaining five subscales; ‘guilt’ (total=22.71), ‘responsibility’ (total=17.21), ‘shame’ (total=24.56), ‘rejection’ (total=40.00) and ‘search for explanation’ (total=28.71). The lowest mean scores for the subscales were from either of the natural modes of death. This suggests that participants who were presented with either of the suicide deaths were more likely to answer with one of the higher responses (4 or 5) and those presented with the natural modes of death were more likely to answer with the lower responses (1 or 2).

Table 1. The Total Scores and Standard Deviations for each subscale and Total GEQ Score for each mode of death.

Subscale	Mode of Death
	Suicide Expected		Suicide Unexpected		Accidental		Natural Expected		Natural Unexpected
	Total	SD	Total	SD	Total	SD	Total	SD	Total	SD
Stigmatization	33.22	4.42	33.10	6.98	29.54	6.34	29.51	5.57	28.94	6.19
Guilt	22.12	4.23	22.71	4.31	20.58	4.38	19.92	3.74	20.42	3.14
Responsibility	16.84	4.10	17.27	4.15	12.50	4.25	11.77	3.69	11.96	4.08
Shame	23.82	3.80	24.56	4.80	21.26	4.60	21.15	4.15	20.68	3.88
Rejection	37.51	7.23	40.00	8.05	25.40	10.04	25.09	7.52	25.26	7.43
Search for Explanation	27.92	3.72	28.71	3.30	27.76	3.34	27.32	3.27	27.52	3.53
Overall GEQ	154.49	20.96	159.38	24.10	131.88	22.98	129.77	20.31	129.98	19.35

Cronbach’s coefficient alpha test was carried out prior to analysis to investigate the reliability of each of the subscales (see Appendix F). Any score for the subscales that was equivalent to 0.7 or higher was viewed as reliable (Cronbach, 1951). Table 2 below shows the Cronbach coefficient alpha scores for the six subscales and the total GEQ score. Most of the subscales had alpha scores of 0.7 or higher, with the highest being 0.93 for rejection. The search for explanation subscale alpha, however, is less than 0.7 at 0.55, suggesting that this subscale should be treated with caution. An attempt was made to increase this alpha score. By removing two items (9 and 44) and having a 5-item scale instead the alpha score increased to 0.64 which was a slight improvement but is still not considered reliable (see Appendix G). The overall GEQ alpha score proved to also be reliable with a score of 0.94.

Table 2. The Cronbach alpha for each subscale and the total GEQ score.

Subscale	No. of items	Alpha
Responsibility	5	.881
Rejection	11	.925
Stigmatization	10	.811
Search for Explanation	7	.548
Guilt	6	.803
Shame	6	.713
Total GEQ Score	44	.942

The data was then analysed via a one-way between participants ANOVA to compare the effect of mode of death on an individual’s perceived grief (see Appendix H). Analysis showed that there was a significant effect for perceived grief in terms of the total GEQ score (F(4,247) = 22.77, p = .0001). The results of this analysis are presented in table 3.

An ANOVA was also conducted to examine the effect of mode of death on perceived grief, in terms of each of the subscales (see Appendix I). The analysis showed that there was a significant effect of mode of death on perceived grief for all the subscales apart from the search for explanation subscale (F(4,247) = 1.19, p = .316). Table 4 shows the results from this analysis.

Table 3. ANOVA analysis for the total GEQ score.

Total GEQ Score		ANOVA
	F(4,247)=	Sig.
Between groups	22.77	.0001

Table 4. ANOVA analysis for each of the subscales.

Subscale		ANOVA
	F(4,247)=	Sig.
Responsibility	22.30	.0001
Rejection	42.06	.0001
Stigmatization	6.32	.0001
Search For Explanation	1.19	.316
Guilt	4.51	.002
Shame	8.54	.0001

The final stage of analysis included the Tukey HSD post-hoc test. The test was carried out on the six subscales (see Appendix J), with only those that were significant after the ANOVA analysis being reported, and the total GEQ (see Appendix K). The Tukey post-hoc was chosen to examine any differences between the subscales for each mode of death. Table 5 below shows that when comparing the suicide expected mode of death with the natural expected mode of death, all the subscales were significant. Analysis also found that all subscales were significant in the accidental mode of death and in the natural unexpected mode of death except the ‘guilt’ subscale (p=.299) and (p=.206) respectively.

Table 5. Tukey HSD post-hoc test comparing suicide expected for each subscale and the overall GEQ with the four other modes of death.

Suicide Expected
	Suicide Unexpected	Accidental	Natural Expected	Natural Expected
Total GEQ	.793	.0001	.0001	.0001
Responsibility	.985	.0001	.0001	.0001
Rejection	.546	.0001	.0001	.0001
Stigmatization	1.000	.018	.014	.003
Guilt	.948	.299	.043	.206
Shame	.910	.023	.013	.002

Table 6 shows the results of the post-hoc analysis between the suicide unexpected mode of death and the four other modes of death. Analysis found that all the subscales were significant in the natural expected and natural unexpected modes of death and all the subscales but ‘guilt’ were significant for the accidental mode of death (p=.066). Both tables 5 and 6 show that there was no significance in the subscales between the two suicide modes of death. This suggests that the anticipatory effect of an expected suicide didn’t impact participants perceived grief reactions within the subscales.

Table 6. Tukey HSD post-hoc test comparing suicide unexpected for each subscale and the overall GEQ with the four other modes of death.

Suicide Unexpected
	Suicide Expected	Accidental	Natural Expected	Natural Expected
Total GEQ	.793	.0001	.0001	.0001
Responsibility	.985	.0001	.0001	.0001
Rejection	.546	.0001	.0001	.0001
Stigmatization	1.000	.027	.022	.006
Guilt	.948	.066	.005	.039
Shame	.910	.0001	.0001	.0001

Discussion

This research set out to explore how the public perceives suicide survivors compared to other survivor groups in order to understand how they may provide support in the future. Negative perceptions of suicide survivors could lead to a lack of social support and so it is important to know if the public’s perception is a negative one. This research also examined whether perceptions of grief differed when the suicide was expected or unexpected to provide researchers with an insight as to whether a suicide being expected impacts the public’s perceptions of survivors.

The results of the current study support the hypotheses that participants, who are exposed to either of the suicide vignettes, will perceive their grief to be more unique and will report higher overall scores on the Grief Experience Questionnaire than those exposed to the accidental, natural expected and natural unexpected modes of death.

Firstly, the highest total GEQ score was for the suicide unexpected mode of death and there was only a slight difference between this score and that of the total GEQ score for the suicide expected mode of death. This suggests that the participant’s perceptions of suicide survivors did not drastically change when the suicide was anticipated. There was, however, a more significant difference between the two suicide deaths and the accidental and two natural deaths suggesting that survivors of suicide are perceived as having more frequent and severe grief reactions than natural death survivors. This result supports those found by Bailley at al. (1999) highlighting that people expect the more unique and higher grief reactions for suicide survivors that were shown by these researchers. Further support for the conclusion that suicide survivors experience a more unique grieving process is found within the current study in that the highest total scores of each of the six subscales were for either of the suicide deaths and the lowest scores were either for the natural expected or the natural unexpected deaths. The current study also contrasts the results found by Cleiren et al. (1994) who suggested that the grief of suicide survivors is no more prolonged or complicated than that of other survivor groups.

Referring to the integrative risk factor model proposed by Stroebe et al. (2006), the results of the current study support the model’s conclusion that the nature of the death impacts a bereaved individual’s grief reactions as further analysis found that the mode of death had a significant impact on participant’s perception of grief. This result is possibly due to the added grief reactions that apply to survivors of suicide; responsibility, rejection, stigmatization, search for explanation, guilt and shame. This result also demonstrates the GEQ’s ability to effectively reveal the differences in grief reactions between survivors of suicide and other death survivors, which has previously been shown in past research (McIntosh, Arnett & Thomas, 1992; Silverman et al., 1995).

The current study produced findings that suggest participants didn’t perceive there to be any differences between survivors of an expected suicide and survivors of an unexpected suicide in terms of the grief reactions within the subscales. A comparison of the two suicide groups found that there was no significance of the six subscales suggesting that being aware that a suicide could possibly occur did not impact participant’s perceptions of the survivors. As a result, the findings of this study cannot relate to previous research suggesting that morbid grief reactions are likely to follow if a death is sudden and unexpected (Wadland et al., 1988). The reason for this could be that the public assumes all suicide survivors must have had a sense that the deceased was at risk of committing suicide and therefore see there to be no relevant differences between the two groups.

The responsibility subscale was found to be highly reliable and provided significant results for the other modes of death when compared with both expected and unexpected suicide. This suggests that participants perceive suicide survivors to somehow be responsible for the death but do not perceive this feeling of responsibility in survivors of accidental, expected natural or unexpected natural deaths. This supports the conclusion reached by Miles and Demi (1992) that survivors of suicide show higher feelings of responsibility and guilt than other mourners. However, although the guilt subscale in the present study was found to be significant overall, it was insignificant when comparing both suicide deaths with accidental death. This suggests that participants perceived survivors of the accidental death to have feelings of guilt. In the accidental vignette, Laura died from falling and hitting her head on the kitchen worktop and so, participants may perceive Michael to have some feelings of guilt for not being there when she fell. The insignificance of this result supports Barrett and Scott’s (1990) findings that guilty reactions are no more heightened for suicide survivors when compared to survivors of other deaths. Further support of their conclusion has been shown in the present study as guilt was also found to be insignificant when comparing the suicide expected mode of death with the natural expected.

The subscale of rejection was also significant overall and proved significant when comparing both suicide deaths with the other three modes of death. This result suggests that feeling rejected by the deceased is perceived to be unique to suicide survivors and not seen as something that would be felt by survivors of other deaths. The perception of this feeling by the participants supports previous research suggesting that feelings of rejection are more intense for those bereaved by suicide (Yufit, 1977). Shame is another subscale that was found to be significant when comparing expected and unexpected suicide to the other modes of death. Participant’s perception that these feelings would be higher in those bereaved by suicide can support previous findings that shame felt by this group of survivors can be so intense it leads to them lying about the cause of death (Range & Calhoun, 1990). Participants may have perceived Michael to feel ashamed and embarrassed that Laura had chosen to commit suicide because of the negative stigmatization that is associated with the act itself.

The stigmatization of suicide is possibly the main reason as to why participants within the current study perceived survivors to experience the feelings depicted in the subscales much more often than the other survivor groups. The stigmatization subscale also gave significant results for the accidental, natural expected and unexpected natural deaths when compared to either of the suicide deaths. The negative perceptions toward suicide survivors presented by the participants in the current study are similar to those shown by participants in preceding research suggesting that the public does hold negative attitudes towards suicide survivors (Allen et al., 1994). As previously mentioned, Feigelman et al. (2009) concluded that suicide survivors were not the only survivor group to experience feelings of stigmatization as it was also found to be present in the survivors of accidental deaths. The results of the present study refute this finding as there was a significant difference between the perceptions of the accidental death survivors and both suicide survivor groups.

The search for explanation subscale produced no significant results. This particular finding was slightly surprising as it has been previously suggested that, following a suicide, searching for an explanation is harder to resolve than it is following other modes of death (Wallace, 1977). The results of this subscale were also surprising because the uncertainty involved in an unexpected suicide is likely to leave survivors questioning why their loved one chose to take their own life. An explanation of this insignificance could be due to the Cronbach coefficient alpha for this subscale being a score that is considered unreliable. This suggests that the items within the subscale are not measuring the same underlying construct and that these results should be considered with caution. The result of this reliability test for the search for explanation subscale differs from previous research that has found the subscale to have high internal consistency reliability (Barrett & Scott, 1989).

Potential Limitations

The first possible limitation for the current study is that it is investigating perceived grief reactions rather than real grief reactions. The use of fictional characters within a hypothetical scenario would not produce as severe grief reactions as a person’s account of actual bereavement and, for this reason, these results cannot be generalised to the reactions experienced by actual survivors. Despite this, using perception studies also has advantages, the first of these being that asking individuals to perceive grief reactions allows for a larger sample size than having to recruit participants that have been bereaved and are willing to participate. Another strength of perception studies is that researchers are able to control what participants are exposed to. This means that variables such as the mode of death and relationship to the deceased can be controlled which is necessary as they have been found to have an impact on grief reactions (Cleiren, 1993; Sanders, 1999).

A further possible limitation of the present study is the lack of demographics obtained from participants. One demographic that wasn’t collected from participants was their gender. Research has suggested that there are significant differences in grief reactions following the loss of a spouse in widows and widowers (Stroebe, Stroebe & Schut, 2001). The results from this research found that widowers showed higher frequencies of grief reactions and therefore future research should consider comparing the effect of either the husband or wife dying within the vignette. A final limitation of the present study is the use of university students as participants. Although participant’s ages were not collected, and it is therefore impossible to know their average age, it is most likely that the majority of participants were between the ages of 18-25 and are therefore unlikely to be married. The limitation here is that it may be difficult for a young student to try and relate to how Michael would feel after losing Laura.

A strength of the current study is the use of the GEQ scale (Barret & Scott, 1989) as it has been deemed an effective tool in differentiating between different modes of death, and the overall scales and subscales within have been found to have high internal consistency. This high internal consistency for the overall scale was also found in the present study with the use of the Cronbach alpha reliability test.

Future Research

A suggestion for future research could be to examine participant’s perception on another group of survivors: those who have been bereaved by a homicide. As research has found that losing a loved one in a violent manner, either accident, suicide or homicide, is a risk factor for the development of complicated grief reactions (Currier, Holland & Neimeyer, 2006), it could be beneficial to see if the public’s perception of homicide survivors differs to their perception of suicide survivors. Another suggestion for future research would be to not only compare perceptions of grief for various modes of death, but to also compare different relationships to the deceased. Research investigating the difference in grief reactions between parents who have lost a child and spouses who have lost their partner has found that bereaved parents experience more intense grief responses than bereaved spouses (Middleton, Raphael, Burnett & Martinek, 1998). Therefore, it would be interesting to investigate whether parents as suicide survivors are perceived any differently to spouses as suicide survivors.

Implications of Findings

Understanding how the public perceives suicide survivors has implications regarding treatment for these bereaved individuals. As the results of this study highlight that people stigmatize suicide and the survivors by association, it will be important to consider how supported the bereaved feel by their social groups. Lack of social support leads to an individual dealing with their grief alone and this can prolong and complicate the process. It has also been suggested that people are unsure of how they can console suicide survivors (Dyregrov et al., 2003) and so treatment for suicide survivors could include counselling for the friends and family of the bereaved on how they can help and support them. This research also suggests that there is a need for the public to have a better understanding of suicide survivors in order to reduce the level of stigmatization associated with it.

Conclusion

In summary, the results of this study show that the mode of death does have a significant effect on participant’s perception of the survivors. More specifically, survivors of suicide are perceived to have more intense and a higher frequency of unique grief reactions than survivors of accidental, natural expected or natural unexpected deaths. With the exception of searching for an explanation, the results suggest that participants perceived the survivors of suicide to experience more feelings of responsibility, rejection, stigmatization, guilt and shame than the other death survivors. The current research provides an understanding of how the public perceive suicide survivors and therefore an indication of how the public might support someone they know who has been bereaved by suicide. Future research could investigate whether these perceptions of differences between survivors is still shown when the bereaved is a parent who has lost a child to suicide.

References

Allen, B. G., Calhoun, L. G., Cann, A., & Tedeschi, R. G. (1994). The effect of cause of death on responses to the bereaved: Suicide compared to accident and natural causes. OMEGA-Journal of Death and Dying, 28(1), 39-48.

Arbuckle, N. W., & de Vries, B. (1995). The long-term effects of later life spousal and parental bereavement on personal functioning. The Gerontologist, 35(5), 637-647.

Bailley, S. E., Dunham, K., & Kral, M. J. (2000). Factor Structure of the Grief Experience Questionnaire (GEQ). Death Studies, 24(8), 721-738.

Bailley, S. E., Kral, M. J., & Dunham, K. (1999). Survivors of suicide do grieve differently: Empirical support for a common sense proposition. Suicide and Life-Threatening Behavior, 29(3), 256-271.

Barrett, T. W., & Scott, T. B. (1989). Development of the Grief Experience Questionnaire. Suicide and Life-Threatening Behavior, 19(2), 201-215.

Barrett, T. W., & Scott, T. B. (1990). Suicide bereavement and recovery patterns compared with non-suicide bereavement patterns. Suicide and Life-Threatening Behavior, 20(1), 1-15.

Battle, A. (1984). Group therapy for survivors of suicide. Crisis: The Journal of Crisis Intervention and Suicide Prevention, 5(1), 45-58.

Bernhardt, G. R., & Praeger, S. G. (1983). After suicide: Meeting the needs of the survivors. Washington, DC: Paper presented at the annual convention of the American Personnel and Guidance Association.

Boelen, P. A., & van den Bout, J. (2005). Complicated grief, depression, and anxiety as distinct postloss syndromes: a confirmatory factor analysis study. American Journal of Psychiatry, 162(11), 2175–2177.

Bonanno, G., Wortman, C., & Nesse, R. (2004). Prospective patterns of resilience and maladjustment during widowhood. Psychology and Aging, 19(2), 260-271.

Calhoun, L. G., Selby, J. W., & Faulstich, M. E. (1980). Reactions to the parents of the child suicide: A study of social impressions. Journal of Consulting and Clinical Psychology, 48(4), 535.

Calhoun, L. G., Selby, J. W., & Selby, L. E. (1982). The psychological aftermath of suicide: An analysis of current evidence. Clinical Psychology Review, 2(3), 409-420.

Callahan, J. (2000). Predictors and correlates of bereavement in suicide support group participants. Suicide and Life-Threatening Behavior, 30(2), 104-124.

Clark, S. E., & Goldney, R. D. (1995). Grief reactions and recovery in a support group for people bereaved by suicide. Crisis, 16(1), 27-33.

Cleiren, M. P. (1993). Bereavement and adaptation: A comparative study of the aftermath of death. Washington, DC: Hemisphere.

Cleiren, M. P., Diekstra, R. F., Kerkhof, A. J., & Van der Wal, J. (1994). Mode of death and kinship in bereavement: Focusing on “who” rather than “how”. Crisis: The Journal of Crisis Intervention and Suicide Prevention, 15(1), 22-36.

Cowles, K. V., & Rodgers, B. L. (1991). The concept of grief: A foundation for nursing research and practice. Research in Nursing and Health, 14(2), 119-127.

Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests. Psychometrika, 16(3), 297-334.

Currier, J. M., Holland, J. M., & Neimeyer, R. A. (2006). Sense-making, grief, and the experience of violent loss: Toward a mediational model. Death studies, 30(5), 403-428.

Demi, A. S. (1984). Social adjustment of widows after a sudden death: Suicide and non-suicide survivors compared. Death Education, 8(s1), 91-111.

Dunn, R. G., & Morrish-Vidners, D. (1988). The psychological and social experience of suicide survivors. OMEGA-Journal of Death and Dying, 18(3), 175-215.

Dyregrov, K., Nordanger, D., & Dyregrov, A. (2003). Predictors of psychosocial distress after suicide, SIDS and accidents. Death Studies, 27(2), 143-165.

Ellenbogen, S., & Gratton, F. (2001). Do they suffer more? Reflections on research comparing suicide survivors to other survivors. Suicide and Life-Threatening Behavior, 31(1), 83-90.

Farberow, N. L., Gallagher-Thompson, D., Gilewski, M., & Thompson, L. (1992). Changes in grief and mental health of bereaved spouses of older suicides. Journal of Gerontology, 47(6), P357-P366.

Feigelman, W., Gorman, B. S., & Jordan, J. R. (2009). Stigmatization and suicide bereavement. Death Studies, 33(7), 591-608.

Fine, C. (1997). No time to say good-bye. New York: Doubleday.

Ginsburg, G. (1971). Public conceptions and attitudes about suicide. Journal of Health and Social Behavior, 12(3), 200-207.

Grad, O. T., & Zavasnik, A. (1996). Similarities and differences in the process of bereavement after suicide and after traffic fatalities in Slovenia. OMEGA- Journal of Death and Dying, 33(3), 243-251.

Harwood, D., Hawton, K., Hope, T., & Jacoby, R. (2002). The grief experiences and needs of bereaved relatives and friends of older people dying through suicide: a descriptive and case-control study. Journal of affective disorders, 72(2), 185-194.

Holland, J. M., & Currier, J. M., & Neimeyer, R. A. (2014). Validation of the integration of stressful life experiences scale – short form in a bereaved sample. Death Studies, 38(4), 234-238.

Holland, J. M., & Neimeyer, R. A. (2011). Separation and Traumatic Distress in Prolonged Grief: The Role of Cause of Death and Relationship to the Deceased. Journal of Psychopathology and Behavioural Assessment, 33(2), 254-263.

Jordan, J. R. (2001). Is suicide bereavement different? A reassessment of the literature. Suicide and Life-Threatening Behavior, 31(1), 91-102.

Kovarsky, R. S. (1989). Loneliness and disturbed grief: a comparison of parents who lost a child to suicide or accidental death. Archives of Psychiatric Nursing, 3(2), 86-89.

Kristensen, P., Weisæth, L., & Heir, T. (2012). Bereavement and mental health after sudden and violent losses: a review. Psychiatry: Interpersonal & Biological Processes, 75(1), 76-97.

Latham, A. E., & Prigerson, H. G. (2004). Suicidality and bereavement: Complicated grief as psychiatric disorder presenting greatest risk for suicidality. Suicide and Life-Threatening Behavior, 34(4), 350-362.

Lightner, C., & Hathaway, N. (1990). Giving sorrow words. New York: Warner Books.

Marris, P. (1958). Widows and their families. London: Routledge & Kegan Paul.

Marris, R. (1981). Pathways to suicide. Baltimore: John Hopkins University Press.

McIntosh, J. L. (1987a). Preface. In E. J. Dunne, J. L. McIntosh, & K. Dunne-Maxim (Eds.), Suicide and its aftermath: Understanding and counseling the survivors (pp. 19-30). New York: Norton.

McIntosh. J. L. (1987b). Research, therapy, and educational needs. In E. J. Dunne, J. L. McIntosh, & K. Dunne-Maxim (Eds.), Suicide and its aftermath: Understanding and counseling the survivors (pp. 263-277). New York: Norton.

McIntosh, J. L. (1993). Control group studies of suicide survivors: A review and critique. Suicide and Life-threatening Behavior, 23(2), 146-161.

McIntosh, J. L., Arnett, E., & Thomas, R. (1992). Grief and bereavement instruments: A comparison. In 25^th annual meeting of the American Association of Suicidology, Chicago.

McIntosh, J. L., & Kelly, L. D. (1992). Survivors’ reactions: Suicide vs. other causes. Crisis: The Journal of Crisis Intervention and Suicide Prevention, 13(2), 82-93.

Middleton, W., Raphael, B., Burnett, P., & Martinek, N. (1998). A longitudinal study comparing bereavement phenomena in recently bereaved spouses, adult children and parents. Australian and New Zealand Journal of Psychiatry, 32(2), 235-241.

Miles, M. S., & Demi, A. S. (1992). A comparison of guilt in bereaved parents whose children died by suicide, accident, or chronic disease. OMEGA-Journal of Death and Dying, 24(3), 203-215.

Office for National Statistics. (2016a,). Deaths registered in England and Wales: 2015. Retrieved from https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/bulletins/deathsregistrationsummarytables/2015.

Office for National Statistics. (2016b,). Suicides in the UK: 2015 registrations. Retrieved from https://www.ons.gov.uk/peoplepopulationandcommunity/birthsdeathsandmarriages/deaths/bulletins/suicidesintheunitedkingdom/2015registrations.

Osterweis, M., Solomon, F., & Green, M. (1984). Reactions to particular types of bereavement. In M. Osterweis, F. Solomon, & M. Green (Eds.), Bereavement: Reactions, Consequences, and Care. Washington, DC: National Academies Press.

Parkes, C. M. (1970). The first year of bereavement. Psychiatry, 33, 442-462.

Parkes, C. M., & Weiss, R. S. (1983). Recovery from bereavement. New York: Basic Books.

Prigerson, H., Ahmed, I., Silverman, G. K., Saxena, A. K., Maciejewski, P. K., Jacobs, S. C., … Hamirani, M. (2002). Rates and risks of complicated grief among psychiatric clinic patients in Karachi, Pakistan. Death Studies, 26(10), 781-792.

Prigerson, H. G., Horowitz, M. J., Jacobs, S. C., Parkes, C. M., Aslan, M., Goodkin, K., … Bonanno, G. (2009). Prolonged grief disorder: Psychometric validation of criteria proposed for DSM-V and ICD-11. PLoS Med, 6(8).

Prigerson, H. G., Maciejewski, P. K., Reynolds, C. F., Bierhals, A. J., Newsom, J. T., Fasiczka, A., … Miller, M. (1995). Inventory of Complicated Grief: a scale to measure maladaptive symptoms of loss. Psychiatry research, 59(1), 65-79.

Rando, T. A. (1993). Treatment of complicated mourning. Champaign, IL: Research Press.

Range, L. M., & Calhoun, L. G. (1990). Responses following suicide and other types of death: The perspective of the bereaved. OMEGA-Journal of Death and Dying, 21(4), 311-320.

Reed, M. D., & Greenwald, J. Y. (1991). Survivor-Victim Status, Attachment, and Sudden Death Bereavement. Suicide and Life-Threatening Behavior, 21(4), 385-401.

Samaritans. (2017). Key trends from the Samaritans Suicide Statistics Report 2017. Retrieved from http://www.samaritans.org/about-us/our-research/facts-and-figures-about-suicide.

Sanders, C. M. (1983). Effects of sudden vs. chronic illness death on bereavement outcome. OMEGA-Journal of Death and Dying, 13(3), 227-241.

Sanders, C. M. (1999). Grief: The Mourning After: Dealing with Adult Bereavement (2^nd ed.). New York: Wiley and Sons.

Schut, H., Stroebe, M., van den Bout, J., & Terheggen, M. (2001). The efficacy of bereavement interventions: Determining who benefits. In M. Stroebe, R. Hansson, W. Stroebe, & H. Schut (Eds.), Handbook of bereavement research: Consequences, coping and care (pp. 705-737). Washington: American Psychological Association.

Schuyler, D. (1973). Counselling suicide survivors: Issues and answers. OMEGA-Journal of Death and Dying, 4(4), 313-321.

Seguin, M., Kiely, M. C., & Lesage, A. (1993). After suicide: a unique mourning experience?. Sante mentale au Quebec, 19(2), 63-82.

Seguin, M., Lesage, A., & Kiely, M. (1995). Parental bereavement and accident: A comparative study. Suicide and Life-Threatening Behavior, 25(4), 489-497.

Servaty-Seib, H. L., & Pistole, M. C. (2007). Adolescent grief: Relationship category and emotional closeness. OMEGA- Journal of Death and Dying, 54(2), 147-167.

Silverman, E., Range, L., & Overholser, J. (1995). Bereavement from suicide as compared to other forms of bereavement. OMEGA-Journal of Death and Dying, 30(1), 41-51.

Solomon, C. G., & Shear, M. K. (2015). Complicated Grief. The New England Journal of Medicine, 372(2), 153-160.

Sprang, G., & McNeil, J. (1995). The many faces of bereavement: The nature and treatment of natural, traumatic, and stigmatised grief. New York: Brunner/Mazel.

Stroebe, M. S., Folkman, S., Hansson, R. O., & Schut, H. (2006). The prediction of bereavement outcome: Development of an integrative risk factor framework. Social Science & Medicine, 63(9), 2440-2451.

Stroebe, W., & Schut, H. (2001). Risk factors in bereavement outcome: A methodological and empirical review. In M. S. Stroebe, R. O. Hansson, W. Stroebe, & H. Schut (Eds.), Handbook of bereavement research: Consequences, coping and care (pp. 349-371). Washington, DC: American Psychological Association.

Stroebe, M., Stroebe, W., & Schut, H. (2001). Gender differences in adjustment to bereavement: An empirical and theoretical review. Review of General Psychology, 5(1), 62-83.

Sveen, C. A., & Walby, F. A. (2008). Suicide survivors’ mental health and grief reactions: A systematic review of controlled studies. Suicide and Life-Threatening Behavior, 38(1), 13-29.

Van der Wal, J. (1990). The aftermath of suicide: A review of empirical evidence. OMEGA-Journal of Death and Dying, 20(2), 149-171.

Van Dongen, C. J. (1990). Agonizing questioning: Experiences of survivors of suicide victims. Nursing Research, 39(4), 224-229.

Van Dongen, C. J. (1993). Social context of postsuicide bereavement. Death suicides, 17(2), 125-141.

Wadland, W. C., Keller, B., Jones, W., & Chapados, J. (1988). Sudden, unexpected death and the role of the family physician. Family Systems Medicine 6(2), 176.

Wallace. S. (1977). On the atypicality of suicide bereavement. In B. Danto & A. Kutscher (Eds.), Suicide and bereavement. New York: Arno Press.

Weiss, R. S. (1999). Grief, Bonds, and Relationships. In M. S. Stroebe, R. O. Hansson, W. Stroebe, & H. Schut (Eds.), Handbook of bereavement research: Consequences, coping and care (pp. 47-62). Washington, DC: American Psychological Association.

Worden, J. W. (1991). Grief counselling and grief therapy (2^nd Ed,). New York: Springer.

Worden, J. W. (2010). Grief counselling and grief therapy (4^th Ed,). London: Routledge.

Yufit, R. (1977). Suicide, bereavement, & time perspective. In B. Danto & A. Kutscher (Eds.), Suicide and bereavement. New York: Arno Press.

Appendices

Appendix A: Vignette – five versions presented online

Please read the following hypothetical scenario.  After reading, please complete the questionnaire which follows:

Michael and Laura have been married for 14 years and live together in a small village just outside of Harrogate. They met each other when they first started high school and were in the same friendship circle. Michael and Laura began their relationship on a school skiing trip to Chamonix when they were 16 and have been together ever since, getting married when they were in their late twenties.

Michael and Laura have 3 children; Erin, Jake and Adam, aged between 5 and 11 years of age.  Erin has just started at the local infant school, and Adam has been at the local high school for the last 3 months having moved up from junior school in September.  The family share the farm cottage where they live with their 2 pet dogs Bilbo and Chipper, along with their cat Luna.

Laura works part-time as a vet and in her spare time volunteers for the RSPCA, and Michael works as a hotel manager at a hotel in Harrogate. Alongside volunteering, Laura’s enjoys baking at home with the children and meets up occasionally with her old university friends for weekend trips to the theatre.  Michael is an avid rugby fan but now watches his son, Adam, play rugby on a Sunday morning, rather than playing rugby himself any more. Adam is also very good at other sports, including football and cricket, and Erin and Jake both enjoy keeping active with swimming and going out on their bikes.

Every Sunday after rugby practice, the children visit their grandparents and Laura and Michael have some time to themselves. They often go out for a pub lunch or a country walk.  On this particular Sunday the couple went shopping for their children’s Christmas presents. When they had finished shopping, Laura went home to hide the presents while Michael drove to his parent’s house to pick up the children. When Michael returned home with the children, he called out for Laura but heard no reply.  The children went upstairs to play, and on entering the kitchen, Michael found Laura dead on the kitchen floor.

Suicide expected: Laura died by taking an intentional overdose of over the counter sleeping tablets.  Laura had been suffering with depression for the past 12 years, she had already attempted suicide once before.

Suicide unexpected: Laura died by taking an intentional overdose of over the counter sleeping tablets.

Accidental death: Laura died from head injuries, caused by an accidental fall to the floor, striking her head on the corner of the kitchen worktop as she fell. 

Natural death expected: Laura had been diagnosed with terminal breast cancer 18 months ago and her health had been deteriorating in recent months.

Natural death unexpected: Laura died from a sudden brain haemorrhage. 

*Next button*

Appendix B: Questionnaire presented online

Please read through the following statements, and answer each question carefully based on the scenario you have just read, identifying how frequently Michael might experience each of the following, from 1 ‘Never’ to 5 ‘Almost Always:

1                2                      3                     4                        5

Never         Rarely         Sometimes         Often         Almost Always

 

Using the scale above, since the death of Laura, in your opinion, how often would Michael:

1. Think people feel uncomfortable offering their condolences to him?

2. Avoid talking about the negative or unpleasant parts of their relationship.

3. Feel like he would never be able to get over Laura’s death.

4. Feel anger or resentment towards Laura after her death.

5. Question why Laura had to die.

6. Find himself unable to stop thinking about how Laura’s death occurred.

7. Think that Laura’s time to die had not yet come.

8. Find himself not accepting that Laura’s death had happened.

9. Try to find a good reason for Laura’s death.

10. Feel avoided by friends?

11. Think that others don’t want to talk to him about Laura’s death?

12. Feel like no one cares to listen to him?

13. Think that his neighbours and in-laws don’t offer enough concern?

14. Feel like a social outcast?

15. Think people were gossiping about him or his wife Laura?

16. Think that people were probably wondering about what kind of personal problems he and Laura had experienced?

17. Feel others may have blamed him for Laura’s death.

18. Think that the death somehow reflected negatively on him or his family.

19. Feel stigmatised by Laura’s death?

20. Think of times before Laura’s death when he could have made her life more pleasant.

21. Wished that he hadn’t said or done certain things during his relationship with Laura.

22. Feel like there was something very important he wanted to make up to Laura.

23. Feel like maybe he didn’t care enough about Laura.

24. Feel somehow guilty after Laura’s death.

25. Feel as though Laura had some kind of complaint against him at the time of her death.

26. Feel as though if he had somehow been a different person, Laura may not have died.

27. Feel as though he made Laura unhappy long before her death.

28. Feel like he missed an early sign which may have indicated to him that Laura was not going to be alive much longer.

29. Feel as though the problems he and Laura had together contributed to her untimely death.

30. Avoid talking about the death of Laura.

31. Feel uncomfortable revealing the cause of Laura’s death.

32. Feel embarrassed about Laura’s death.

33. Feel uncomfortable about meeting someone who knew him and Laura.

34. Not mention Laura’s death to people he met casually.

35. Feel as though Laura chose to leave him.

36. Feel deserted by Laura.

37. Feel as though the death was somehow a deliberate abandonment of him.

38. Feel that Laura never considered what the death might do to him.

39. Sense some feeling that Laura rejected him by dying.

40. Wonder about Laura’s motivation for not living longer.

41. Feel that Laura was somehow getting even with him by dying.

42. Feel as though he should have somehow prevented Laura’s death.

43. Tell someone that the cause of Laura’s death was something than what it really was.

44. Feel that Laura’s death was a senseless and wasteful loss of life.

*Next button*

Appendix C: Ethics Proforma approved by BPS standards

Psychology Research Project

Research and Ethics Proforma

Student Name:    Supervisor Name:  

Title of Project: Suicide bereavement: a unique bereavement experience?

Project Code:  

Description of Methods

In the space below, briefly and simply describe the main research question of the study, your rationale for asking this question, and the methods that you will use.  The purpose of this section is to demonstrate that you know why you are doing this study and what you will be doing.

This study will be investigating perceived grief following different modes of death; suicide, natural and accidental. Previous studies (Bailey, Kral & Dunham, 1999) have suggested that the grief experienced by suicide survivors is unique compared to the grief experiences following other modes of death.

The independent variable for this study will be the mode of death, which will have 5 levels; ‘accidental’, ‘expected suicide’, ‘unexpected suicide’, ‘expected natural’ and ‘unexpected natural’. The dependent variable will be the participants perceived grief which will be measured through the answers they give during the questionnaire. This study is particularly interested in investigating the differences in perceived grief when comparing suicide to other modes of death.

In this study, perceived grief will be assessed using a vignette study followed by an online questionnaire. Participants will be recruited via the SONA system and will therefore consist of undergraduate university students. Each participant will read the same story regarding an individual, their life and how they have died. The only difference will be the last sentence of the vignette which will include one of five deaths; ‘accidental death’, ‘expected suicide’, ‘unexpected suicide’, ‘expected natural’ and ‘unexpected natural’. Participants will be randomly given one of the scenarios but will be unaware that there is more than one.

Once participants have read the vignettes they will be provided with an online questionnaire on Qualtrics to measure their perceived grief. The questionnaire given to participants will be adapted from the GEQ (Barrett & Scott, 1989) to measure perceived grief rather than personal experience. The data will then be analysed using an ANOVA with post-HOC tests.

Ethical Issues

In the space below, briefly discuss the key ethical issues that relate to your project (one short paragraph per issue) and how you intend to deal with these. A non-exhaustive list of issues you may wish to consider includes: informed consent, vulnerable participants, right to withdraw, anonymity, confidentiality, deception, debriefing, data storage.

Informed consent: Prior to completing the questionnaire participants will be given a detailed information sheet (included in appendices) providing them with details of the study and informing them what will be required of them. Due to the sensitive nature of the study, an exclusion criterion will be included within the information sheet preventing any distress to potential participants who may have experienced recent bereavement. As the questionnaire will be online, participants will be made aware that when submitting their answers they are giving their consent automatically.

Right to withdraw:  Due to the data being collected anonymously, participants will not be able to withdraw their data once they have submitted their answers to the questionnaire online. This will be made aware to the participants in the information sheet where they will also be told that after reading the details in the information sheet they do not have to continue their participation in the study if they do not wish.

Anonymity: As this study involves an online questionnaire, participants will be responding anonymously and will not be required to provide any personal details.

Confidentiality: All data collected during this study will be kept confidential. The data will then be stored on a password protected computer which only I will have access to. If the data wherever to be included in any scholarly activity, it would still remain in an anonymous form with all personal details being kept confidential.

Deception: Participants will only be slightly deceived as they will not be made aware that the vignette they are given is one of five possible scenarios. However, they will be informed of this in the debrief sheet (included in appendices) provided to them upon completion of the questionnaire.

Debriefing: The debrief sheet will be given to the participants once they have completed the study and will include information regarding the aim and contact details for myself and my research supervisor. As the study is focusing on a sensitive subject, participants will be provided with contact details for further support if they feel they would need it.

Data storage: As previously stated, collected data will be stored on a password protected computer that only I will be able to access and if the data are ever used in a published paper, it will be stored on a research archive within the university for up to 5 years. After these 5 years all data will be destroyed.

Action Plan

In the table below, list the specific actions that you need to take (or have taken) and when you will take them to progress with your project. Pay particular attention to actions related to the ethics of your project.

What	When
Write information sheet, debrief sheet and vignette.	Done
Attend Qualtrics and ‘Thinking Like a Researcher’ workshop.	15th November 2016
Have finalised questionnaire and vignette completed.	22nd  November 2016
Create advertisement for study.	End of November 2016
Read handbook for Online Research Participation System and set up SONA.	End of November 2016
Start recruiting participants and collecting data.	Mid December 2016-January 2017
Complete Lit Review and write up.	December 2016
Attend quantitative analysis lecture and workshop and any other relevant workshops.	January-  early February 2017
Finish data collection and enter SPSS data.	Mid February 2017
Finish analysis.	End of February 2017
Write up methods and results sections.	Mid March 2017
Write discussion.	Early April 2017
Print out and proofread.	Early April 2017
Print out two copies and have them bound.	Mid April 2017
Submit dissertation.	28th April 2017

Study Materials and Ethics Documents

List each of the measures, questionnaires, and stimuli sets you will be using.  In the appendices, include any unpublished measures in full, along with the information sheet, consent form, and debrief sheet (where applicable). Where possible, your materials should be fully in place before your supervisor can pass your Research and Ethics Proforma.

Unpublished materials

Information Sheet – see Appendix A

Debrief Sheet – see Appendix B

Vignette – see Appendix C

References

Bailey, S.E., Kral, M.J., & Dunham, K. (1999). Survivors of Suicide do Grieve Differently: Empirical support for a common sense proposition. Suicide and Life-Threatening Behaviour, 29, 256-271.

Barrett, T.W., & Scott, T.B. (1989). Development of the grief experience questionnaire. Suicide and Life-Threatening Behaviour, 19, 201-215.

Risk Assessment

1. Will the proposed data collection take place solely online, on campus, or at your own residence?

 Yes  (Please proceed to question 6)

 No, it will take place somewhere else (Please complete all questions)

2. Where will the data collection take place? (Tick as many as apply if data collection will take place in multiple venues)

 Residence of participant

 School

 Business/Voluntary Organisation

 Public Venue (e.g., Youth Club, Church, etc.)

 Other (Please specify) _____________________________

How will you ensure your personal safety while at the research venue?

3. How will you travel to and from the data collection venue?

 On foot

 By car

 Public Transport

 Other (Please specify) ______________________________

How will you ensure your personal safety when travelling to/from the data collection venue?

5. Whenever you go to collect data, you must ensure that someone you trust knows where you are going (without breaching the confidentiality of your participants), how you are getting there (preferably including your travel route), when you expect to get back, and what to do should you not return at the specified time. Please outline here the procedure you propose using to do this:

6. Are you aware of any potential risks to your health and wellbeing associated with the venue where the research will take place and/or the research topic?

 Yes (Please outline below)

 No

7. Does this research project require a health and safety risk analysis for the procedures to be used?

 Yes

 No

If yes, what is the current status of the health and safety risk assessment?

Confirmation of Ethical Abidance by Student

*Next button*

Appendix E: Cronbach alpha SPSS Output

Total GEQ Score – Overall Scale Reliability

Reliability Statistics
Cronbach’s Alpha	Cronbach’s Alpha Based on Standardized Items	N of Items
.942	.937	44

Responsibility     Rejection

Reliability Statistics
Cronbach’s Alpha	Cronbach’s Alpha Based on Standardized Items	N of Items
.881	.882	5

Reliability Statistics
Cronbach’s Alpha	Cronbach’s Alpha Based on Standardized Items	N of Items
.925	.923	11

Stigmatization    Search for Explanation

Reliability Statistics
Cronbach’s Alpha	Cronbach’s Alpha Based on Standardized Items	N of Items
.811	.813	10

Reliability Statistics
Cronbach’s Alpha	Cronbach’s Alpha Based on Standardized Items	N of Items
.548	.592	7

Reliability Statistics
Cronbach’s Alpha	Cronbach’s Alpha Based on Standardized Items	N of Items
.713	.701	7

        Guilt        Shame

Reliability Statistics
Cronbach’s Alpha	Cronbach’s Alpha Based on Standardized Items	N of Items
.803	.811	6

Appendix G: Cronbach alpha SPSS Output – Search for Explanation with Items 9 and 44 removed

Search for Explanation

Reliability Statistics
Cronbach’s Alpha	Cronbach’s Alpha Based on Standardized Items	N of Items
.639	.642	5

Appendix H: ANOVA SPSS Output

ANOVA
TotalGEQscore
	Sum of Squares	df	Mean Square	F	Sig.
Between Groups	42394.398	4	10598.600	22.774	.000
Within Groups	114951.538	247	465.391
Total	157345.937	251

Appendix I: ANOVA SPSS Output

ANOVA
		Sum of Squares	df	Mean Square	F	Sig.
Responsibility	Between Groups	1508.878	4	377.220	22.995	.000
	Within Groups	4051.927	247	16.405
	Total	5560.806	251
AbandonmentRejection	Between Groups	11059.710	4	2764.927	42.056	.000
	Within Groups	16238.893	247	65.745
	Total	27298.603	251
Stigmatization	Between Groups	892.737	4	223.184	6.315	.000
	Within Groups	8729.592	247	35.342
	Total	9622.329	251
SearchForExplanation	Between Groups	56.250	4	14.062	1.190	.316
	Within Groups	2918.750	247	11.817
	Total	2975.000	251
Guilt	Between Groups	286.668	4	71.667	4.514	.002
	Within Groups	3921.269	247	15.876
	Total	4207.937	251
ShameEmbarrassment	Between Groups	619.134	4	154.784	8.544	.000
	Within Groups	4474.517	247	18.115
	Total	5093.651	251

Appendix J: Tukey HSD Post-hoc test SPSS Output

Multiple Comparisons
Tukey HSD
Dependent Variable	(I) Modeofdeath	(J) Modeofdeath	Mean Difference (I-J)	Std. Error	Sig.	95% Confidence Interval
						Lower Bound	Upper Bound
Responsibility	Suicide expected	Suicide Unexpected	-.42770	.81451	.985	-2.6660	1.8106
		Accidental	4.34314*	.80607	.000	2.1280	6.5582
		Natural death expected	5.06955*	.79447	.000	2.8863	7.2528
		Natural death unexpected	4.88314*	.80607	.000	2.6680	7.0982
	Suicide Unexpected	Suicide expected	.42770	.81451	.985	-1.8106	2.6660
		Accidental	4.77083*	.81845	.000	2.5217	7.0200
		Natural death expected	5.49725*	.80702	.000	3.2795	7.7150
		Natural death unexpected	5.31083*	.81845	.000	3.0617	7.5600
	Accidental	Suicide expected	-4.34314*	.80607	.000	-6.5582	-2.1280
		Suicide Unexpected	-4.77083*	.81845	.000	-7.0200	-2.5217
		Natural death expected	.72642	.79851	.893	-1.4679	2.9207
		Natural death unexpected	.54000	.81005	.963	-1.6860	2.7660
	Natural death expected	Suicide expected	-5.06955*	.79447	.000	-7.2528	-2.8863
		Suicide Unexpected	-5.49725*	.80702	.000	-7.7150	-3.2795
		Accidental	-.72642	.79851	.893	-2.9207	1.4679
		Natural death unexpected	-.18642	.79851	.999	-2.3807	2.0079
	Natural death unexpected	Suicide expected	-4.88314*	.80607	.000	-7.0982	-2.6680
		Suicide Unexpected	-5.31083*	.81845	.000	-7.5600	-3.0617
		Accidental	-.54000	.81005	.963	-2.7660	1.6860
		Natural death expected	.18642	.79851	.999	-2.0079	2.3807
AbandonmentRejection	Suicide expected	Suicide Unexpected	-2.49020	1.63058	.546	-6.9711	1.9907
		Accidental	12.10980*	1.61369	.000	7.6753	16.5443
		Natural death expected	12.41546*	1.59046	.000	8.0448	16.7861
		Natural death unexpected	12.24980*	1.61369	.000	7.8153	16.6843
	Suicide Unexpected	Suicide expected	2.49020	1.63058	.546	-1.9907	6.9711
		Accidental	14.60000*	1.63846	.000	10.0974	19.1026
		Natural death expected	14.90566*	1.61559	.000	10.4660	19.3454
		Natural death unexpected	14.74000*	1.63846	.000	10.2374	19.2426
	Accidental	Suicide expected	-12.10980*	1.61369	.000	-16.5443	-7.6753
		Suicide Unexpected	-14.60000*	1.63846	.000	-19.1026	-10.0974
		Natural death expected	.30566	1.59855	1.000	-4.0872	4.6985
		Natural death unexpected	.14000	1.62166	1.000	-4.3164	4.5964
	Natural death expected	Suicide expected	-12.41546*	1.59046	.000	-16.7861	-8.0448
		Suicide Unexpected	-14.90566*	1.61559	.000	-19.3454	-10.4660
		Accidental	-.30566	1.59855	1.000	-4.6985	4.0872
		Natural death unexpected	-.16566	1.59855	1.000	-4.5585	4.2272
	Natural death unexpected	Suicide expected	-12.24980*	1.61369	.000	-16.6843	-7.8153
		Suicide Unexpected	-14.74000*	1.63846	.000	-19.2426	-10.2374
		Accidental	-.14000	1.62166	1.000	-4.5964	4.3164
		Natural death expected	.16566	1.59855	1.000	-4.2272	4.5585
Stigmatization	Suicide expected	Suicide Unexpected	.11152	1.19553	1.000	-3.1738	3.3969
		Accidental	3.67569*	1.18315	.018	.4244	6.9270
		Natural death expected	3.70625*	1.16612	.014	.5017	6.9108
		Natural death unexpected	4.27569*	1.18315	.003	1.0244	7.5270
	Suicide Unexpected	Suicide expected	-.11152	1.19553	1.000	-3.3969	3.1738
		Accidental	3.56417*	1.20131	.027	.2629	6.8654
		Natural death expected	3.59473*	1.18454	.022	.3396	6.8499
		Natural death unexpected	4.16417*	1.20131	.006	.8629	7.4654
	Accidental	Suicide expected	-3.67569*	1.18315	.018	-6.9270	-.4244
		Suicide Unexpected	-3.56417*	1.20131	.027	-6.8654	-.2629
		Natural death expected	.03057	1.17204	1.000	-3.1903	3.2514
		Natural death unexpected	.60000	1.18899	.987	-2.6674	3.8674
	Natural death expected	Suicide expected	-3.70625*	1.16612	.014	-6.9108	-.5017
		Suicide Unexpected	-3.59473*	1.18454	.022	-6.8499	-.3396
		Accidental	-.03057	1.17204	1.000	-3.2514	3.1903
		Natural death unexpected	.56943	1.17204	.989	-2.6514	3.7903
	Natural death unexpected	Suicide expected	-4.27569*	1.18315	.003	-7.5270	-1.0244
		Suicide Unexpected	-4.16417*	1.20131	.006	-7.4654	-.8629
		Accidental	-.60000	1.18899	.987	-3.8674	2.6674
		Natural death expected	-.56943	1.17204	.989	-3.7903	2.6514
SearchForExplanation	Suicide expected	Suicide Unexpected	-.78676	.69129	.786	-2.6865	1.1129
		Accidental	.16157	.68413	.999	-1.7185	2.0416
		Natural death expected	.60081	.67429	.900	-1.2521	2.4538
		Natural death unexpected	.40157	.68413	.977	-1.4785	2.2816
	Suicide Unexpected	Suicide expected	.78676	.69129	.786	-1.1129	2.6865
		Accidental	.94833	.69464	.651	-.9606	2.8572
		Natural death expected	1.38758	.68494	.257	-.4947	3.2698
		Natural death unexpected	1.18833	.69464	.429	-.7206	3.0972
	Accidental	Suicide expected	-.16157	.68413	.999	-2.0416	1.7185
		Suicide Unexpected	-.94833	.69464	.651	-2.8572	.9606
		Natural death expected	.43925	.67771	.967	-1.4231	2.3016
		Natural death unexpected	.24000	.68751	.997	-1.6493	2.1293
	Natural death expected	Suicide expected	-.60081	.67429	.900	-2.4538	1.2521
		Suicide Unexpected	-1.38758	.68494	.257	-3.2698	.4947
		Accidental	-.43925	.67771	.967	-2.3016	1.4231
		Natural death unexpected	-.19925	.67771	.998	-2.0616	1.6631
	Natural death unexpected	Suicide expected	-.40157	.68413	.977	-2.2816	1.4785
		Suicide Unexpected	-1.18833	.69464	.429	-3.0972	.7206
		Accidental	-.24000	.68751	.997	-2.1293	1.6493
		Natural death expected	.19925	.67771	.998	-1.6631	2.0616
Guilt	Suicide expected	Suicide Unexpected	-.59069	.80127	.948	-2.7926	1.6112
		Accidental	1.53765	.79297	.299	-.6415	3.7167
		Natural death expected	2.19312*	.78155	.043	.0454	4.3409
		Natural death unexpected	1.69765	.79297	.206	-.4815	3.8767
	Suicide Unexpected	Suicide expected	.59069	.80127	.948	-1.6112	2.7926
		Accidental	2.12833	.80514	.066	-.0842	4.3409
		Natural death expected	2.78381*	.79390	.005	.6021	4.9655
		Natural death unexpected	2.28833*	.80514	.039	.0758	4.5009
	Accidental	Suicide expected	-1.53765	.79297	.299	-3.7167	.6415
		Suicide Unexpected	-2.12833	.80514	.066	-4.3409	.0842
		Natural death expected	.65547	.78553	.920	-1.5032	2.8141
		Natural death unexpected	.16000	.79688	1.000	-2.0299	2.3499
	Natural death expected	Suicide expected	-2.19312*	.78155	.043	-4.3409	-.0454
		Suicide Unexpected	-2.78381*	.79390	.005	-4.9655	-.6021
		Accidental	-.65547	.78553	.920	-2.8141	1.5032
		Natural death unexpected	-.49547	.78553	.970	-2.6541	1.6632
	Natural death unexpected	Suicide expected	-1.69765	.79297	.206	-3.8767	.4815
		Suicide Unexpected	-2.28833*	.80514	.039	-4.5009	-.0758
		Accidental	-.16000	.79688	1.000	-2.3499	2.0299
		Natural death expected	.49547	.78553	.970	-1.6632	2.6541
ShameEmbarrassment	Suicide expected	Suicide Unexpected	-.73897	.85593	.910	-3.0911	1.6131
		Accidental	2.56353*	.84706	.023	.2358	4.8913
		Natural death expected	2.67259*	.83487	.013	.3783	4.9668
		Natural death unexpected	3.14353*	.84706	.002	.8158	5.4713
	Suicide Unexpected	Suicide expected	.73897	.85593	.910	-1.6131	3.0911
		Accidental	3.30250*	.86007	.001	.9390	5.6660
		Natural death expected	3.41156*	.84806	.001	1.0811	5.7421
		Natural death unexpected	3.88250*	.86007	.000	1.5190	6.2460
	Accidental	Suicide expected	-2.56353*	.84706	.023	-4.8913	-.2358
		Suicide Unexpected	-3.30250*	.86007	.001	-5.6660	-.9390
		Natural death expected	.10906	.83911	1.000	-2.1969	2.4150
		Natural death unexpected	.58000	.85125	.960	-1.7593	2.9193
	Natural death expected	Suicide expected	-2.67259*	.83487	.013	-4.9668	-.3783
		Suicide Unexpected	-3.41156*	.84806	.001	-5.7421	-1.0811
		Accidental	-.10906	.83911	1.000	-2.4150	2.1969
		Natural death unexpected	.47094	.83911	.980	-1.8350	2.7769
	Natural death unexpected	Suicide expected	-3.14353*	.84706	.002	-5.4713	-.8158
		Suicide Unexpected	-3.88250*	.86007	.000	-6.2460	-1.5190
		Accidental	-.58000	.85125	.960	-2.9193	1.7593
		Natural death expected	-.47094	.83911	.980	-2.7769	1.8350
*. The mean difference is significant at the 0.05 level.

Appendix K: Tukey HSD Post-hoc test SPSS Output

Multiple Comparisons
Dependent Variable:   TotalGEQscore
Tukey HSD
(I) Modeofdeath	(J) Modeofdeath	Mean Difference (I-J)	Std. Error	Sig.	95% Confidence Interval
					Lower Bound	Upper Bound
Suicide expected	Suicide Unexpected	-4.88480	4.33831	.793	-16.8066	7.0370
	Accidental	22.61020*	4.29338	.000	10.8118	34.4086
	Natural death expected	24.71661*	4.23158	.000	13.0881	36.3451
	Natural death unexpected	24.51020*	4.29338	.000	12.7118	36.3086
Suicide Unexpected	Suicide expected	4.88480	4.33831	.793	-7.0370	16.8066
	Accidental	27.49500*	4.35930	.000	15.5155	39.4745
	Natural death expected	29.60142*	4.29844	.000	17.7891	41.4137
	Natural death unexpected	29.39500*	4.35930	.000	17.4155	41.3745
Accidental	Suicide expected	-22.61020*	4.29338	.000	-34.4086	-10.8118
	Suicide Unexpected	-27.49500*	4.35930	.000	-39.4745	-15.5155
	Natural death expected	2.10642	4.25309	.988	-9.5812	13.7941
	Natural death unexpected	1.90000	4.31458	.992	-9.9566	13.7566
Natural death expected	Suicide expected	-24.71661*	4.23158	.000	-36.3451	-13.0881
	Suicide Unexpected	-29.60142*	4.29844	.000	-41.4137	-17.7891
	Accidental	-2.10642	4.25309	.988	-13.7941	9.5812
	Natural death unexpected	-.20642	4.25309	1.000	-11.8941	11.4812
Natural death unexpected	Suicide expected	-24.51020*	4.29338	.000	-36.3086	-12.7118
	Suicide Unexpected	-29.39500*	4.35930	.000	-41.3745	-17.4155
	Accidental	-1.90000	4.31458	.992	-13.7566	9.9566
	Natural death expected	.20642	4.25309	1.000	-11.4812	11.8941
*. The mean difference is significant at the 0.05 level.