Elderspeak is a patronizing style of speech that is commonly used when conversing with older adults (Whitbourne, 2008). It is characterized by features such as simplified vocabulary and sentence structure; exaggerated intonation and vocal pitch; slowed rate of speech; use of personal terms of endearment (e.g., sweetie); and collective pronoun usage (Kemper, Finter-Urczyk, Ferrell, Harden, & Billington, 1998). A reasonable question to ask is, “Why do younger people use elderspeak?” This is a complex question that may be better understood if one instead asks, “Under what circumstances is elderspeak more or less likely to be used?” Existing theoretical and empirical work has uncovered contextual variables that are more or less likely to prompt the use of elderspeak.
A fundamental component of interpersonal communication is an active evaluation and adaptive process in which all participants engage. This process is explicated in detail as part of communication accommodation theory, which asserts that a speaker will change or alter his or her speech based on the speaker’s evaluation of the other participants (Coupland, Coupland, Giles, & Henwood, 1988). Accommodations for participants in discourse are not always beneficial and may hinder communication when they are excessive or based on inaccurate assumptions (Hummert, Garstka, Ryan, & Bonnesen, 2004). This phenomenon has been labeled the Communication Predicament of Aging (CPA) model, which states that a speaker’s communication style or accommodations for the listener are based on incorrect assumptions, such as incompetence or loss of functioning (Ryan, Hummert, & Boich, 1995). Elderspeak is one kind of excessive form of accommodation that a speaker may make when interacting with an older adult and may be especially relevant for professionals working in long-term care facilities. For example, elder-speak has been shown to be prevalent in long-term care facilities, with one study finding that 22% of sentences between caregivers and residents were categorized as having the characteristics of elderspeak (Caporael, 1981).
Empirical research has uncovered various factors that alter the use or acceptability of elderspeak. Several studies indicate that elder-speak may be considered more acceptable and more likely to occur in institutional settings such as hospitals or nursing homes (Caporael, 1981; Hummert, Shaner, Garstka, & Henry, 1998; LaTourette & Meeks, 2000). Research also suggests that characteristics of older adult recipients, such as cognitive impairment (LaTourette & Meeks, 2000) or depressed mood (Hummert et al., 1998), can alter the likelihood that elderspeak is used.
The phenomenon of elderspeak is problematic in many ways. Elderspeak is judged to be condescending, disrespectful, and unwelcomed by individuals receiving it (Draper, 2005). Furthermore, recipients of elderspeak are evaluated as being less competent, less capable, in a worse mood, and having poorer memories and communication skills compared to recipients of non-patronizing speech (Draper, 2005; LaTourette & Meeks, 2000). Studies have also generally found that nurses who engage in elder-speak are viewed more negatively with respect to a number of traits (e.g., professionalism, respectfulness, friendliness, helpfulness, competence) compared to nurses using neutral speech (Balsis & Carpenter, 2005; Ryan, Bourhis, & Knops, 1991). Finally, mounting evidence indicates that elder-speak may increase the likelihood of residents in institutional settings resisting care (Williams, Herman, Gajewski, & Wilson, 2008).
The existing theoretical and empirical literature described above identifies a number of factors that may increase the likelihood that elderspeak is used by individuals working in long-term care facilities. Because elderspeak can have many negative outcomes, it is important to better understand when professional caregivers may perceive elderspeak to be more or less appropriate. Consequently, the purpose of the current study was to assess how appropriate professional caregivers perceived elder-speak to be in response to a variety of resident-related and situational variables.
Participants were 134 certified nursing assistants (CNAs) employed at long-term care facilities located in the midwestern United States. Of the participants who responded to demographic items, the majority were women (n = 119; men, n = 3), ranging in age from 19 to 71 (mean = 37.35, SD = 15.65 years), and were primarily from the Midwest (n = 108; other, n = 13). The majority of participants reported their ethnicity as Caucasian (n = 90), followed by African American (n = 4), Hispanic/Latino (n = 3), Asian (n = 2), and Asian/Caucasian (n = 1). Participants were employed as CNAs from 2 to 516 months (mean = 119.42 months, SD = 123.71 months). Participants reported skilled nursing (n = 81), assisted living (n = 17), memory care (n = 13), or a combination of units (n = 9) as their primary units of work.
Data were collected both onsite (n = 127) and electronically (n = 7). The researchers visited and collected data at nine long-term care facilities in the midwestern United States. Three facilities were categorized as nursing homes, three were categorized as assisted living facilities, and three facilities had both a nursing home and assisted living component.
The questionnaire used for the current study comprised two components. The first component was a demographic questionnaire that included 10 questions intended to identify characteristics of the participants (e.g., gender, age). The second component included 36 items that described various cues/variables, and participants were asked to rate each of the items on a scale from 1 to 4 (1 being not at all appropriate, 4 being extremely appropriate) based on their perceptions of the appropriateness of elderspeak in response to that particular cue/variable. Items for the second component of the questionnaire were developed by a team of researchers based on several sources of information: (a) existing empirical and theoretical literature on “old-age” cues that evoke elderspeak (Ryan et al., 1995); (b) incorporating additional old-age cues that had not been investigated in the current research literature, but that could potentially affect the use of elderspeak (e.g., the presence of hearing or visual deficits); (c) some of the team members’ direct experience working in long-term care settings; and (d) brainstorming sessions among the research team members about other variables that could potentially determine whether elderspeak is used.
In addition, questionnaire items were created such that there would be a corresponding item (or items) that was rationally/theoretically related to it, and these sets of related items comprised a subscale. This was done so that items related to each other could be statistically compared to test hypotheses about the perceived appropriateness of a variable. For example, the item describing a cognitively intact resident had three corresponding items: (a) one that described a mildly impaired individual, (b) one that described a severely impaired individual, and (c) one that described a resident who appeared confused or disoriented. These four items comprised the “cognitive status” sub-scale. Planned contrasts were then run to test the hypothesis (based on existing literature) that elderspeak would be perceived as being less appropriate with residents showing no cognitive impairment compared to residents with some degree of cognitive impairment.
A total of 36 items were developed and categorized into 14 sub-scales as described above. Nine subscales concerned characteristics of the resident, namely the resident’s gender, age, visual functioning, auditory functioning, ability to ambulate independently, emotional status, cognitive status, cooperativeness during care, and level of education. Two factors concerned characteristics of the relationship between the CNA and the resident, namely frequency of interactions and familiarity with the resident. Two factors concerned characteristics of the job being performed by the CNA, such as the type of task the CNA was completing with the resident and the workload of the CNA. The final factor concerned the characteristics of the present situation, specifically whether others were present during a CNA–resident interaction. Internal consistency of the measure was calculated using Cronbach’s alpha and was found to be 0.977.
Data collection sessions were conducted onsite (except for the data collected electronically) at the aforementioned long-term care facilities. After signing a consent form (approved by an institutional review board) that outlined their rights as participants in this research, participants were provided instructions regarding the questionnaire. Instructions included a brief description of the purpose of the questionnaire, written examples of what elderspeak entails, an audio clip demonstrating elderspeak, and examples of how to complete the questionnaire items. Instructions were written such that participants were not being asked about their own use of elderspeak, but their general perceptions of how elderspeak is perceived by “a typical nursing assistant.” This was done to minimize the influence of social desirability on responding. Following the instructions, participants completed the 36 appropriateness ratings.
As mentioned previously, the survey comprised 14 subscales, with each subscale containing two to five items. A series of 14 repeated measures analysis of variance (ANOVAs) was conducted to compare mean ratings of perceived appropriateness among all items in that particular subscale. If results of an ANOVA revealed that the Mauchly’s Test of Sphericity was violated, a Greenhouse-Geisser correction was used to correct degrees of freedom. Because multiple main effect analyses were conducted, a Bonferroni correction was computed, resulting in a critical alpha level of 0.026.
For subscales with more than two items, if a significant main effect was found, planned contrasts were conducted to determine whether significant differences existed between specific item pairs within that subscale that were hypothesized to differ from each other. There were a total of four subscales where significant main effects were found and that included more than two items. Because multiple planned contrasts were completed for these four sub-scales, a Bonferroni correction was calculated for each subscale, resulting in critical alpha levels ranging from 0.043 to 0.048.
Table 1 displays the means and standard deviations for all questionnaire items. Overall, ratings for the perceived appropriateness of elderspeak were relatively low, with means ranging from 1.44 to 2.23. On average, the lowest appropriateness ratings concerned items involving the presence of others during an interaction, lack of familiarity with the resident, younger residents, well-educated residents, and residents who are cognitively intact. The highest appropriateness ratings included items involving familiarity with the resident (i.e., working with them regularly, knowing their name), a resident displaying positive or sad affect, interactions that occur during hands-on tasks, and a resident displaying confusion or disorientation.
Means and Standard Deviations for the 36 Items Describing Various Cues/Variables and the Appropriateness of Elderspeak
Of the 14 repeated measures ANOVAs conducted, eight produced nonsignificant findings (Table 2). Results demonstrated that there were no significant differences between the items in the following subscales: the presence/absence of visual impairment, the presence/absence of hearing impairment, the resident’s ability/inability to ambulate independently, gender, how (un)cooperative a resident is during personal care tasks, amount of education an individual obtained or the prestige of a resident’s previous occupation, whether or not the CNA was busy, and how familiar the CNA was with the resident (i.e., could the CNA recall the resident’s name?).
Summary of Subscale Analyses and Planned Contrast Results
Six repeated measures ANOVAs produced significant findings (Table 2). One such variable was age. Findings indicated that elder-speak was perceived as more appropriate when residents are older (compared to younger) than 70, F(1, 122) = 13.06, p < 0.001.
The analysis regarding the perceived appropriateness of elder-speak in response to a resident’s emotional state also produced a significant main effect, F(2, 240) = 14.94, p < 0.001. Planned contrasts revealed significant differences that indicate elderspeak is perceived to be less appropriate when older adults are angry compared to when they appear happy, F(1, 120) = 18.11, p < 0.001 or sad/depressed, F(1, 120) = 24.67, p < 0.001.
Differences between the perceived appropriateness of elder-speak in relation to a resident’s cognitive status were examined. Sphericity was violated within this analysis; therefore, degrees of freedom were corrected using the Greenhouse-Geisser estimates of sphericity (ε= 0.763). A significant main effect was found for this subscale, F(2.29, 276.86) = 7.87, p = 0.002. Results of all three planned contrasts revealed significant differences between the items, indicating that the use of elderspeak is perceived as being more appropriate when residents have mild memory problems, F(1, 121) = 9.48, p = 0.001; severe memory problems, F(1, 121) = 6.66, p = 0.011; or are disoriented/confused, F(1, 121) = 15.22, p < 0.001 compared to when a resident has no cognitive impairment.
A significant main effect was found for the relationship subscale that concerned the frequency of interactions between a CNA and a resident, F(1.52, 180.34) = 23.39, p < 0.001. Sphericity was violated within this analysis; therefore, degrees of freedom were corrected using the Greenhouse-Geisser estimates of sphericity (ε= 0.758). Planned contrasts revealed significant differences, suggesting that when CNAs and residents have regular interactions, elderspeak is more likely to be perceived as appropriate compared to when there have been no previous interactions between a CNA and a resident, F(1, 119) = 50.74, p < 0.001, or when there have been infrequent interactions, F(1, 119) = 43.71, p = 0.017.
The situational factors subscale included five items that reflected whether an interaction between a CNA and a resident occurred when no one else was present or if various other people (e.g., supervisor, family members) were present. Sphericity was violated within this analysis; therefore, degrees of freedom were corrected using the Greenhouse-Geisser estimates of sphericity (ε= 0.900). A significant main effect was found for this subscale, F(3.60, 428.18) = 4.28, p = 0.011. Although four planned contrasts were completed, only two revealed significant differences, suggesting that elderspeak is perceived as being less appropriate when other residents, F(1, 119) = 5.01, p = 0.027, or family members, F(1, 119) = 13.09, p < 0.001, are present during a CNA–resident interaction. Nonsignificant results were found when comparing the item describing that no one else was present during an interaction to items reflecting that a supervisor, F(1, 119) = 3.65, p = 0.058, or coworker F(1, 119) = 3.68, p = 0.057, was present (not shown).
Finally, the perceived appropriateness of elderspeak was examined in relation to the type of task the CNA was engaged in with the resident. Results suggest that there is a significant difference in the perceived appropriateness of elderspeak with regard to task type, such that if the CNA–resident interaction is during a hands-on care-giving situation (e.g., dressing), elderspeak is rated as more appropriate versus when the CNA–resident interaction is not during a hands-on caregiving situation, F(1, 120) = 8.25, p = 0.005.
The purpose of this study was to evaluate contextual variables that may be related to the perceived appropriateness of elderspeak by CNAs. Results indicated that higher ratings of appropriateness were associated with the following: situations involving older (i.e., >70) versus younger residents; residents who were cognitively impaired versus those without cognitive impairment; those who were displaying sad or happy affect as opposed to anger; situations where the interaction with a resident occurs with no one else present; when the CNA had regular and consistent contact with a resident; and when the CNA–resident interaction occurred during hands-on caregiving tasks. It should be noted that most mean ratings were <2, which indicates that under most circumstances staff consider elderspeak to be at most “somewhat” appropriate. This is encouraging given the general consensus in the literature that elderspeak has mostly negative consequences and should generally be avoided (Draper, 2005).
The findings that the appropriateness of elderspeak is related to age, the presence of cognitive impairment, and mood are supported by theory and previous research (Hummert et al., 1998; Kemper, Anagnopoulos, Lyons, & Heberlein, 1994; Ryan et al., 1995). Findings that elderspeak is perceived as more appropriate with cognitively impaired individuals suggest that CNAs may believe that elderspeak is an effective means for improving communication (i.e., residents will better understand them if elderspeak is used). The finding that appropriateness ratings were lower when the resident was described as angry indicates that CNAs may have some awareness that elder-speak can be perceived as disrespectful and could antagonize an already agitated resident.
Other variables that were found to be related to perceived appropriateness of elderspeak, such as the type of task being completed with the resident and the absence of others during an interaction, have not to our knowledge been explored in other studies. The result that elderspeak is rated as more appropriate during hands-on tasks may suggest that CNAs assume elderspeak conveys a sense of caring/nurturance during intimate care situations or that it serves to calm potentially agitated residents. Unfortunately, previous research shows these assumptions are incorrect in that individuals who use elderspeak are rated as being less caring or nurturing and that elder-speak can increase the likelihood of resistance to care in those older adults with dementia (Ryan et al., 1991; Williams et al., 2008). The finding that elderspeak is generally perceived to be less appropriate when others are present during a CNA–resident interaction is particularly interesting and seems to suggest that CNAs have some level of awareness that elderspeak could be perceived negatively by others.
The findings regarding the regularity of contact with residents seem to contradict theoretical predictions in that the CPA model hypothesizes that elderspeak is more likely to occur in situations where a younger person is interacting with an unfamiliar conversational partner. One explanation for our apparent contrary findings may be that as staff interact with residents over an extended period of time, they may perceive a level of intimacy where speech similar to elderspeak is deemed as more appropriate and less likely to be misinterpreted as disrespectful.
Limitations and Future Directions
Several limitations of this study must be acknowledged. First, the sample obtained for this study was predominantly Caucasian women who grew up and lived in the midwestern United States. The restricted diversity in the sample limits the generalizability of the findings to other populations. In fact, the role of cultural differences and how they affect the use and impact of elderspeak is generally understudied and an important area for future research. It is possible that what are considered disrespectful linguistic accommodations in one culture are not always considered so in another culture (Giles & Gasiorek, 2010). As an example, terms of endearment such as “sweetie” may be considered more acceptable and within cultural norms in the southeastern United States compared to the upper Midwest where this study was conducted. Therefore, cultural norms may determine the effect of elderspeak and other forms of patronizing speech on older adults. These matters become more important and complex given that approximately 50% of CNAs nationwide are from minority groups (Squillace et al., 2009), which sets the stage for potential communication challenges between staff and residents coming from different cultural backgrounds. One strategy for navigating these complex social interactions is for staff members to actively seek feedback from the older adults they serve about the appropriateness or impact of specific speech accommodations (i.e., the use of a nickname or higher pitched voice intonation) that could be labeled as elderspeak (Ryan et al., 1995). This strategy has the advantage of conveying the respect that staff want to convey as well as reducing the likelihood of unsatisfying interactions.
Two additional limitations are inherent in the instruments used. The questionnaire used in this study included a list of contextual variables presented in isolation of one another. However, real-world interactions with residents involve a combination of variables that occur together, some of which may increase or decrease the perceived appropriateness of elderspeak. The instrument used in this study may need to be modified for future studies to include items involving multiple cues related to higher appropriateness ratings.
Finally, the methodology used in this study limits the types of conclusions that can be made about the actual use of elderspeak. This study inquired about the perceived appropriateness of elderspeak, but self-reports of perceived appropriateness of elderspeak cannot necessarily be equated with actual use of elderspeak in real-life situations. Future research that involves observing a variety of actual CNA–resident interactions may prove useful in further determining what factors are associated with the use of elderspeak.
Implications and Conclusions
The quality of relationships with staff is important to residents living in long-term care facilities and is related to a resident’s quality of life (Grau, Chandler, & Saunders, 1995). CNAs spend the most time with residents, providing 90% of hands-on care and devoting an average of more than 2 hours per day with residents (Kash, Castle, Naufal, & Hawes, 2006). The quality of communication used by CNAs during these interactions can have important effects on residents. For example, the use of elderspeak may increase resistance to care (Williams et al., 2008), threaten self-esteem and well-being, and increase dependency among residents (Ryan, Giles, Bartolucci, & Henwood, 1986).
Given that CNAs spend a great deal of time with residents and these interactions can have a substantial effect on resident functioning, CNAs are an important target for communication-enhancement programs that are designed to reduce the use of elderspeak. Previous research shows that training programs focused on decreasing the use of elderspeak can be effective (Williams, Kemper, & Hummert, 2003). The findings of the current study could have important implications for improving these communication training programs in that a set of contextual “risk factors” were identified that may increase the likelihood that elderspeak is used. Because it is common for younger individuals to be unaware they are using elderspeak, the findings from this study may supplement existing educational programs by enhancing awareness of how certain contextual cues may unknowingly evoke elderspeak. For example, staff may be unaware that elderspeak may be more likely to occur during hands-on caregiving tasks, and that elder-speak increases the probability of the resident engaging in resistance to care during these activities. Furthermore, training programs that involve role-playing or videos could incorporate the contextual variables identified in this study into the training situations to highlight the role these variables play in terms of evoking elderspeak. Increased awareness of the contextual risk factors identified in this study can potentially help CNAs be better prepared for high-risk situations (e.g., hands-on caregiving tasks) so more effective methods of communication can be practiced and used. This kind of training may also improve CNAs’ ability to catch themselves engaging in elderspeak in high-risk situations and make immediate adjustments in their communication style. Of course, the ultimate goal of CNA communication training is to reduce the use of elderspeak to prevent its potential negative effect on residents and improve the quality of care provided.
- Balsis, S. & Carpenter, B.D. (2005). Evaluations of elderspeak in a caregiving context. Clinical Gerontologist, 29, 79–96. doi:10.1300/J018v29n01_07 [CrossRef]
- Caporael, L.R. (1981). The paralanguage of caregiving: Baby talk to the institutionalized aged. Journal of Personality and Social Psychology, 40, 876–884. doi:10.1037/0022-35126.96.36.1996 [CrossRef]
- Coupland, N., Coupland, J., Giles, H. & Henwood, K. (1988). Accommodating the elderly: Invoking and extending a theory. Language in Society, 17, 1–41. doi:10.1017/S0047404500012574 [CrossRef]
- Draper, P. (2005). Patronizing speech to older patients: A literature review. Reviews in Clinical Gerontology, 15, 273–279. doi:10.1017/S0959259806001869 [CrossRef]
- Giles, H. & Gasiorek, J. (2010). Intergenerational communication practices. In Schaie, K.W. & Willis, S.L. (Eds.), Handbook of the psychology of aging (7th ed., pp. 233–247). St. Louis, MO: Elsevier.
- Grau, L., Chandler, B. & Saunders, C. (1995). Nursing homes residents’ perceptions of the quality of their care. Journal of Psychosocial Nursing and Mental Health Services, 33(5), 34–41.
- Hummert, M.L., Garstka, T.A., Ryan, E.B. & Bonnesen, J.L. (2004). The role of age stereotypes in interpersonal communication. In Nussbaum, J.F. & Coupland, J. (Eds.), Handbook of communication and aging research (2nd ed., pp. 91–114). New York, NY: Routledge.
- Hummert, M.L., Shaner, J.L., Garstka, T.A. & Henry, C. (1998). Communication with older adults: The influence of age stereotypes, context, and communicator age. Human Communication Research, 25, 124–151. doi:10.1111/j.1468-2958.1998.tb00439.x [CrossRef]
- Kash, B.A., Castle, N.G., Naufal, G.S. & Hawes, C. (2006). Effect of staff turnover on staffing: A closer look at registered nurses, licensed vocational nurses, and certified nursing assistants. The Gerontologist, 46, 609–619. doi:10.1093/geront/46.5.609 [CrossRef]
- Kemper, S., Anagnopoulos, C., Lyons, K. & Heberlein, W. (1994). Speech accommodations to dementia. Journal of Gerontology, 49, P223–P229. doi:10.1093/geronj/49.5.P223 [CrossRef]
- Kemper, S., Finter-Urczyk, A., Ferrell, P., Harden, T. & Billington, C. (1998). Using elderspeak with older adults. Discourse Processes, 25, 55–73. doi:10.1080/01638539809545020 [CrossRef]
- LaTourette, T.R. & Meeks, S. (2000). Perceptions of patronizing speech by older women in nursing homes and in the community: Impact of cognitive ability and place of residence. Journal of Language and Social Psychology, 19, 463–473. doi:10.1177/0261927X00019004004 [CrossRef]
- Ryan, E.B., Bourhis, R.Y. & Knops, U. (1991). Evaluative perceptions of patronizing speech addressed to elders. Psychology and Aging, 6, 442–450. doi:10.1037/0882-79188.8.131.522 [CrossRef]
- Ryan, E.B., Giles, H., Bartolucci, G. & Henwood, K. (1986). Psycholinguistic and psychosocial components of communication by and with the elderly. Language and Communication, 6, 1–24. doi:10.1016/0271-5309(86)90002-9 [CrossRef]
- Ryan, E.B., Hummert, M.L. & Boich, L.H. (1995). Communication predicaments of aging: Patronizing behavior toward older adults. Journal of Language and Social Psychology, 14, 144–166. doi:10.1177/0261927X95141008 [CrossRef]
- Squillace, M.R., Remsburg, R.E., Harris-Kojetin, L.D., Bercovitz, A., Rosenoff, E. & Han, B. (2009). The National Nursing Assistant Survey: Improving the evidence base for policy initiatives to strengthen the certified nursing assistant workforce. The Gerontologist, 49, 185–197. doi:10.1093/geront/gnp024 [CrossRef]
- Whitbourne, S.K. (2008). Adult development and aging: Biopsychosocial perspectives. Hoboken, NJ: John Wiley & Sons Inc.
- Williams, K., Kemper, S. & Hummert, M.L. (2003). Improving nursing home communication: An intervention to reduce elderspeak. The Gerontologist, 43, 242–247. doi:10.1093/geront/43.2.242 [CrossRef]
- Williams, K.N., Herman, R., Gajewski, B. & Wilson, K. (2008). Elderspeak communication: Impact on dementia care. American Journal of Alzheimer’s Disease & Other Dementias, 24, 11–20. doi:10.1177/1533317508318472 [CrossRef]
Means and Standard Deviations for the 36 Items Describing Various Cues/Variables and the Appropriateness of Elderspeak
|If there are family members present
|If the CNA has never interacted with the resident previously
|If the resident is around other residents
|If the resident is well educated
|If the resident is younger than 70
|If the resident has full mental capacity
|If a supervisor is present
|If coworkers are present
|If the resident previously held what most people consider a prestigious job
|If the CNA is not busy
|If the CNA interacts with the resident infrequently
|If the resident is male
|If the resident appears angry
|If the resident has visual impairment
|If the CNA has a heavy workload
|If the resident is hearing impaired
|If the resident is not hearing impaired
|If the resident has little education
|If the resident has good vision
|If the CNA cannot recall the resident’s name
|If the resident is alone
|If the resident is able to ambulate on his or her own
|If the resident is female
|If the CNA-resident interaction is not during a hands-on caregiving situation
|If the resident is being uncooperative
|If the resident has severe memory problems
|If the resident is older than 70
|If the resident is cooperating
|If the resident has mild memory problems
|If the resident requires assistance with ambulation
|If the resident appears disoriented/confused
|If the CNA knows the resident’s name
|If the CNA–resident interaction is during a hands-on caregiving situation
|If the resident appears happy
|If the resident appears sad/depressed
|If the CNA interacts with the resident on a regular basis
Summary of Subscale Analyses and Planned Contrast Results
||Main Effect Analysis
||Significant Planned Contrasts
||F(1, 123) = 0.31, p = 0.581
||F(1, 119) = 0.00, p = 1.00
|Ability to ambulate
||F(1, 124) = 3.63, p = 0.059
||F(1, 126) = 1.25, p = 0.265
||F(1, 124) = 0.93, p = 0.337
|Educational and occupational background
||F(1.79, 218.03) = 2.40, p = 0.145a
||F(1, 117) = 0.56, p = 0.456
|Relationship: ability to recall resident’s name
||F(1, 124) = 4.18, p = 0.043
||F(1, 122) = 13.06, p < 0.001*
||F(2, 240) = 14.94, p < 0.001*
||Resident appears angry versus resident appears sad/depressed, F(1, 120) = 24.67, p < 0.001*
Resident appears angry versus resident appears happy, F(1, 120) = 18.11, p < 0.001*
||F(2.29, 276.86) = 7.87, p = 0.002*a
||Full mental capacity versus severe memory problems, F(1, 121) = 6.66, p = 0.011*
Full mental capacity versus mild memory problems, F(1, 121) = 9.48, p = 0.001*
Full mental capacity versus appears disoriented or confused, F(1, 121) = 15.22, p < 0.001*
|Relationship: frequency of CNA–resident interactions
||F(1.52, 180.34) = 23.39, p < 0.001*a
||Regular interactions versus no interaction, F(1, 119) = 50.74, p < 0.001*
Regular interactions versus infrequent interactions, F(1, 119) = 43.71, p = 0.017*
|Situational factors: presence of others during a CNA–resident interaction
||F(3.60, 428.18) = 4.28, p = 0.011*a
||Alone versus other residents present, F(1, 119) = 5.01, p = 0.027*
Alone versus family member present, F(1, 119) = 13.09, p < 0.001*
|Type of caregiving task being completed
||F(1, 120) = 8.25, p = 0.005*