Shaking Confidence in Technology: Effects of an Earthquake-Induced Nuclear Disaster on Technology Adoption in Middle-Aged and Older Adults

Abstract

Using the coincidental timing of a national survey conducted in Japan before and after the Fukushima Daiichi nuclear disaster in 2011, this study reports a rare natural experiment that explored how the experience of a nuclear disaster influenced technology adoption in middle-aged and older adults. We conducted path analyses assessing how technology or nontechnology adoption intention and behavior changed before and after the nuclear disaster and whether age could moderate the potential change over and above other relevant factors. Our models supported that Japanese middle-aged to older adults reported fewer technology adoption behaviors after experiencing of the earthquake. However, the negative impact of the earthquake was not more pronounced in older adults. Our results suggest that researchers need to pay more attention to the issue of how loss of trust and/or perceived risk affect technology adoption interacting with other relevant factors, particularly, age-related factors and abilities.

Keywords

technology adoption disaster path analysis

Given its ubiquity in the modern world, technology adoption may benefit older adults by supporting independent living (Czaja et al., 2006). However, in general, older adults reported less comfort in using new technology and were less likely to accept it than younger adults (Czaja et al., 2006; Porter & Donthu, 2006). For instance, in 2019, only 53% of US adults age 65 and above own smartphones compared to 96% of those aged 18 to 29 (https://www.pewinternet.org/fact-sheet/mobile/). Previous studies have demonstrated that adoption of technology in older adults is not a purely technical issue but rather a complex issue that requires considerations of multiple aspects, such as demographic factors, cognitive abilities, and relevant psychological factors (Chung et al., 2010; Czaja et al., 2006; see C. Lee & Coughlin, 2015; Schulz et al., 2014, for reviews).

Particularly, the adoption of new technology can significantly be explained by attitudes toward or perception of technology. The technology acceptance model (TAM; Davis, 1989) is one of the early frameworks that has been effectively used and extended to explain the adoption patterns of different types of new technology. TAM suggested that a user’s intentions to adopt a technology are the single best predictor of use behavior (Davis & Venkatesh, 2004). The Unified Theory of Acceptance and Use of Technology (UTAUT) model (Venkatesh et al., 2012) is another theoretical framework suggesting that people’s intentions to use a new technology and actual technology acceptance are influenced by four key constructs: performance expectancy (usefulness), effort expectancy (ease of use), social influence, and facilitating conditions (Venkatesh et al., 2012). Based on the models, researchers have also attempted to find how social context (Kulviwat et al., 2009; Venkatesh et al., 2003) and individual differences (Chen & Chan, 2014; Porter & Donthu, 2006; Sarker & Wells, 2003), such as age, income, cultural background, technology, and self-efficacy, affect the behavioral intentions and actual acceptance of new technology. The models were recently extended to account for technology acceptance in older adults by adding age-related health and ability factors (e.g., Chen & Chan, 2014; Merkel et al., 2016).

Natural disasters can trigger technological accidents involving breakdown in human-made systems, which can cause catastrophic events, such as the release of hazardous materials or blackouts in large areas (Cruz et al., 2006). Recently, the conjoint natural and technological disaster has been defined as natech disaster and the social and psychological impacts of the disaster have been well documented (Rodriguez et al., 2018; Silver & Garfin, 2016). The Great East Japan Earthquake that occurred in 2011 triggered the meltdown of the Fukushima Daiichi Nuclear Power Plant, and its aftermath had vast consequences for Japan’s society and economy, including more than 15,000 causalities (Kingston, 2012). In particular, since the Great East Japan Earthquake, a number of studies investigated its influence on various aspects of people’s attitudes and behaviors, such as happiness, consumerism, prosocial behavior, and attitudes toward new technology (Goebel et al., 2015; Oishi et al., 2017, 2018; Rehdanz et al., 2015; Tiefenbach & Kohlbacher, 2015a, 2015b). There are also several studies that showed a negative impact of the accident on attitudes toward nuclear power (i.e., increase in perceived risk on nuclear power) (Huang et al., 2013; Siegrist & Visschers, 2013).

It is possible that the experience of the nuclear disaster would increase perceived risk to use advanced technology as perceived risk for nuclear power increased after the disaster (Huang et al., 2013; Siegrist & Visschers, 2013). The failure of the nuclear reactor is associated with multiple failure points in high-technology systems. More specifically, Tokyo Electric Power Company (TEPCO), which ran the reactor, also provides electrical services to citizens, and electrical failures were a common problem following the earthquake, particularly in the very busy subway systems (including Tokyo). Thus, witnessing the failure of the nuclear plant and its aftermath might lead to increase in perceived risk or a temporary loss of trust in the use of advanced technology systems. Japanese adults might also become less likely to adopt or purchase technology products after the disaster because consumerism could be regarded as frivolous after they had witnessed many deaths and suffering (Oishi et al., 2018).

However, although the social and psychological impacts of a nuclear power plant accident have been well documented, there was no study that directly investigated the effect of technological disaster on people’s general attitudes toward technology within the context of major theoretical frameworks explaining technology adoption, such as TAM and UTAUT. Furthermore, given that vulnerability to disaster increases with age-related cognitive and physical declines (Mayhorn, 2005; Ngo, 2001), the potential loss of trust in the use of new technology may be more pronounced in older adults, which will make them more reluctant to adopt technology in spite of its direct benefit to their independent living.

This study was thus designed to explore how technology adoption in Japanese adults changed before and after the disaster and if the potential effect of experiencing the technological disaster was mediated by attitudes toward technology. Particularly, our sample, consisting of middle-aged to older adults, provides an ideal opportunity to study if experiencing a technological disaster would affect older adults’ attitudes toward and uptake of advanced technology. Exploiting a natural experiment approach (Dunning, 2012), we generated path models assessing the effects of timing of completing the national survey on technology adoption after controlling for relevant factors, such as financial status (that impacts purchasing power) and gender. This study also investigated if attitudes toward technology could mediate the potential changes in actual adoption behavior before and after the disaster as well as evaluating any age moderation on the effect.

Given that perceived risk plays a critical role in the development of trust for advanced technology (Hoff & Bashir, 2015) and nuclear power (Huang et al., 2013), we predicted a decrease in technology use behavior after experiencing the earthquake after controlling for gender and financial status at the moment (Hypothesis 1). Based on TAM and UTAUT suggesting a critical role of behavioral intention in technology acceptance, we then expected that the possible negative effect of the earthquake on technology use would be mediated by behavioral intentions (Hypothesis 2). Given that older adults are less likely to accept advanced technology than younger adults, also reporting less comfort in using it (Czaja et al., 2006; Czaja & Sharit, 1998; Tacken et al., 2005), we anticipated that the negative association between the experience of the earthquake and technology adoption would be more pronounced for older adults (Hypothesis 3).

To further validate the possible effect of the disaster on technology adoption, we additionally conducted the same mediation analysis after replacing technology adoption intention and behavior with nontechnology-related consumption intention and behavior (i.e., a control model). It is possible that the effect of the technological disaster would vary for nontechnology-related consumption behavior, given the assumption that the disaster would primarily influence technology-related adoption behavior. Thus, in the control model analysis, we explored whether experiencing the earthquake would still affect nontechnology-related intentions and consumption behavior.

Method

Participants and National Survey

The survey was originally designed to examine older consumers’ behavior and attitudes toward new technology in Japan and was conducted by a well-established Tokyo-based market and public-opinion survey firm (Cheron & Kohlbacher, 2018; Wang et al., 2018). The survey was nationwide by accessing the Japanese household registration data through local administrations. Based on a random sampling procedure, 1,575 Japanese middle-aged to older adults (age range = 40–96 years, mean age = 61.03 years, SD = 11.90; 762 men, 813 women) completed surveys out of 5,000 mailed surveys with a response rate of 31.5%.

Pre-postcard notifications were initially sent to increase the response rate. Then the survey questionnaires were mailed on February 24, 2011, and responses were collected between March 1 and May 23, 2011. A reminder postcard was sent if the survey had not been returned by April 27. The surveys returned to the company on March 15, 2011, or before (N = 792) were considered as pre-earthquake responses. Roughly, the other half (N = 783) returned after March 15 were considered as postearthquake responses.

It should be noted that our intention is not to measure the effects of disaster on technology adoption in Japanese who actually suffered financial losses or injuries due to the earthquake and tsunami, which were locally limited in their impact, but rather to investigate how the indirect experience of nuclear meltdown in Fukushima (e.g., via media) would affect respondents’ attitudes toward and uptake of technology (e.g., Tiefenbach & Kohlbacher, 2015a). This study thus excluded all respondents from the three disaster-affected areas (Iwate, Miyagi, and Fukushima) in the analyses. After excluding the respondents from the three directly hit areas, the sample size used in the following analyses were N = 755 (before the disaster: 375 men, 380 women, age range = 40–96 years, mean age = 61.3) and N = 767 (after the disaster: 356 men, 411 women, age range = 40–92 years, mean age = 60.8). The two samples did not significantly differ in age and household income. The descriptive statistics of two samples without the three areas were reported in Table 1.

Table 1.

Demographics in the Two Samples.

Variables	Sample		t-value
Variables	Before Disaster	After Disaster	t-value
Age	61.3 (11.68)	60.77 (12.07)	0.84
Income^a	3.16 (1.74)	3.03 (1.73)	1.40
Education^b	2.49 (1.02)	2.34 (1.00)	2.88*
Technology
Intention	3.06 (1.27)	2.92 (1.22)	2.04*
Behavior	3.44 (1.99)	3.12 (2.02)	2.90*
Nontechnology
Intention	4.43 (1.28)	4.27 (1.32)	2.25*
Behavior	1.90 (0.63)	1.83 (0.60)	2.22*

1 = less than 2 million yen, 2 = 2~4, 3 = 4~6, 4 = 6~8, 5 = 8~10, 6 = 10~12, 7 = 12~15, 8 = 15~20, 9 = more than 20 million yen. ^b1 = middle school, 2 = high school, 3 = some college, 4 = bachelor degree, 5 = graduate school.

p < .05.

Variables of Interest

Technology adoption intention

The survey included 12 items probing attitudes toward new products and technology. They were adapted from global consumer innovativeness (e.g., I am eager to buy new products as soon as they come out) and technological anxiety (e.g., I feel apprehensive about using technology) measures using a 7-point scale (1 = strongly disagree to 7 = strongly agree) (Meuter et al., 2005; Tellis et al., 2009). The items are shown in the Appendix. As this specific set of items has not been factor analyzed, to identify underlying constructs, an exploratory factor analysis (with varimax rotation) was conducted. The optimal number of factors was determined using the Scree test on eigenvalues greater than 1. There were three factors. Each factor consisted of four items, and the factor loadings were also consistently high across all item (.77–.88). Given that behavioral intention has been considered as a key construct underlying the theoretical mechanism that drives technology adoption (Venkatesh et al., 2012), among these three factors, we selected Factor 2 because the four items involved estimating people’s positive attitudes toward new technology products (e.g., I enjoy the novelty of owning new products). It was named Willingness to Adopt New Technology and used as a behavioral intention index measure in the following path models predicting technology adoption. The eigenvalues of each factor and factor loadings are summarized in Table 2. Cronbach’s alpha for the four items was .87.

Table 2.

Factor Loadings and Eigenvalues of Each Factor for Attitudes Toward Technology.

Item number	Factor loading
Item number	Factor 1Technology anxiety	Factor 2Willingness to adopt	Factor 3Negative perception
1	−.18	.82	.13
2	−.16	.85	.14
3	.14	.11	.65
4	.06	.02	.69
5	.08	.11	.70
6	.20	.09	.65
7	−.10	.81	.06
8	−.10	.85	.05
9	.77	−.08	.21
10	.80	−.12	.23
11	.88	−.19	.07
12	.88	−.18	.09
Eigenvalue	3.82	2.82	1.14

Note. Factor loadings more than .60 appear in bold.

Technology adoption behavior

From the survey, we selected consumer behavior items assessing if respondents ever used 11 different new technology products, such as smartphone and high-end television. They were also adapted from the adoption of innovation measure in Tellis et al. (2009). The items were originally based on the 4-point scale (1 = never seen, 2 = seen but never bought, 3 = bought but never used, and 4 = bought and have used). We recoded the items into dichotomous responses (1, 2, 3 in the original scale into 0 = never used, 4 in the original scale into 1 = used) and calculated a sum of the 11 items as an index of technology adoption. Cronbach’s alpha for the 11 items was .78.

Nontechnology-related intention

The survey had attitudinal items on six different categories of products: home appliances, automobile, cosmetics, food and groceries, sporting goods, and finances, asking if they “are more or less eager to purchase these products and services, relative to high-tech products.” The items were also based on a 7-point scale. An exploratory factor analysis (with varimax rotation) with the six items was also conducted to identify and select items reflecting behavioral intention to purchase nontechnology products.

Based on the factor analysis, we chose the three items on home appliances, cosmetics, and food and groceries, then calculated an average of them as a measure of intention to purchase nontechnology-related products. The factor loadings of the three items were .71, .82, .86, respectively, and an eigenvalue of the factor was 2.46. Cronbach’s alpha for the three items was .74.

Nontechnology-related consumption behavior

The survey also included several items originally designed to measure consumption-coping behavior (E. Lee et al., 2001; Mathur et al., 2008). The items asked if respondents engaged in consumption-related activities for various categories, such as religious and cultural activities, drugs and alcohol, diet and exercise, shopping and eating out, within the past 12 months on the 4-point scale (1 = decreased/stopped, 2 = no change, 3 = increased/started, and 4 = never done). Among the items, we selected the item on “shopping and eating out” as a measure of nontechnology-related consumption behavior. We treated the responses of “never done” as missing values.

Control variables

In line with previous models explaining technology adoption, among various demographics, we chose gender and current financial status as control variables (Porter & Donthu, 2006; Venkatesh et al., 2003, 2012). It should be noted that we used current financial status instead of education or income level (Porter & Donthu, 2006) based on the assumption that current financial status would be more directly associated with intention to purchase and with purchase behavior—compared to education or income level.

Path Analyses: Mediation, Moderated Mediation, and Hypotheses

To test our hypotheses, we used a two-step approach: (a) mediation model and (b) moderated mediation model (Hayes, 2015). To assess the effect of the Fukushima earthquake on happiness and mediation of donation behavior, Tiefenbach and Kohlbacher (2015a) generated a “postearthquake” time dummy variable and used it as a main predictor in their mediation analysis. Similarly, we generated a time dummy variable (i.e., pre-earthquake responses: coded “0” vs. postearthquake responses: coded “1”) based on the return dates as described before. As a first step, in the mediation model, we assessed the effect of responding postearthquake on technology adoption behavior and if the potential effect would be mediated by behavioral intention to adopt new technology after controlling for gender (coded 0 for male and 1 for female) and current financial status (5-point scale: not at well wealthy—not wealthy—neutral—fairly wealthy—very wealthy). In other words, as shown in Figure 1(a), the mediation model tested the direct effect of the postearthquake time dummy variable on the number of technology adoptions (path c) as well as the indirect effect mediated by behavioral intention (path a—b). As a second step, as illustrated in Figure 1(b), the moderated mediation model tested whether the possible effects of earthquake on technology adoption would be moderated by age (path d & e) after controlling for gender and the current financial status.

Figure 1.

Conceptual model of (a) mediation and (b) moderated mediation analyses.

All of these mediation and moderated mediation models were tested using IBM SPSS Statistics for Windows (Version 23.0.) by employing the SPSS macro PROCESS (Hayes, 2013). To estimate indirect effects, we used a bootstrapping method (N = 5,000) as well as a significance test (Sobel, 1982) via PROCESS. Descriptive statistics of demographics and variables of interests are provided in Table 1.

Results

Path Analyses

Effects on technology adoption

As shown in Figure 2(a), the mediation model predicting technology adoption behavior showed that there was a significant direct effect of experiencing the earthquake on the number of technology adoptions, b = −.21, p = .04 (H1), after controlling for gender and current financial status. The negative path coefficient between the earthquake and the number of technology adoptions indicated that the respondents who completed the survey after the earthquake reported fewer technology adoptions. Although, the direct path between the willingness to adopt new technology and the number of technology adoptions was significant, b = .52, p < .01, the indirect path from the earthquake effect to the number of adoptions via the willingness to adopt was not significant, b = −.04, 95% CIs = [−.12, .02], z = −1.30, p = .19 (H2). The result indicates that the willingness to adopt new technology did not mediate the relationship between the earthquake and the number of technology adoptions. In fact, the direct path between the earthquake and the willingness to adopt new technology was not significant either, p = .19. The mediation model explains 12% (R² = .12) of the variance in technology adoption behavior.

Figure 2.

Results of mediation analysis.

In the moderated mediation model (Figure 3), an interaction of experiencing the earthquake with age was not significant both for the willingness to adopt index and the number of technology adoptions, p = .18, 82, respectively. The results indicate that age did not moderate the negative effects of the earthquake (H3). The model explains 17% (R² = .17) of the variance in technology adoption behavior.

Figure 3.

Results of moderated mediation analysis.

Effects on nontechnology-related behavior (control model)

Compared to the mediation model described before, as shown in Figure 2(b), the direct path between the earthquake and nontechnology-related consumption behavior was not significant, p = .15, after controlling for gender and current financial status. However, the indirect path from the earthquake via the intention to purchase was significant, b = −.01, 95% CIs = [−.02, −.003], z = −2.19, p = .03. Whereas, there was no direct relationship between the earthquake and the willingness to adopt new technology products, the direct path between the earthquake and the intention to purchase nontechnology-related products was significant, b = −.17, p < .01, indicating that the respondents who completed the survey after the earthquake reported a lower level of intention to purchase nontechnology-related products. The model explains 1% (R² = .01) of the variance in nontechnology adoption behavior.

The zero-order correlations among variables are provided in Table 3. The unstandardized path coefficients of the predictors in mediation and moderated mediation analyses are summarized in Table 4.

Table 3.

Zero-Order Correlations Among Variables.

Variables	Correlations
Variables	M (SD)	1	2	3	4	5	6	7	8
1. Postearthquake time dummy		—
2. Technology adoption behavior	3.28 (2.01)	−.08**	—
3. Technology adoption intention	2.99 (1.24)	−.05*	.32**	—
4. Nontechnology consumption behavior	1.86 (0.62)	−.06*	.07*	.10**	—
5. Nontechnology intention	4.35 (1.30)	−.06*	.15**	.09**	.11**	—
6. Age	61.03 (11.87)	−.02	−.27**	−.16**	−.01	−.12**	—
7. Gender		.03	−.05	−.16**	.02	.40**	−.06*	—
8. Financial status	2.73 (0.88)	−.07**	.13**	.06*	.24**	.02	.14**	.01	—

*p < .05. **p < .01 two-tailed.

Table 4.

Unstandardized Path Coefficients in Mediation and Moderated Mediation Analyses.

Predictor	Behavioral intention	Behavior
Predictor	Behavioral intention	Direct path	Indirect path via intention
Technology adoption intention and behavior
Mediation model
Postearthquake	−0.09	−.21* (H1)	−.04 (H2)
Behavioral intention		.51**
Controls: gender	−0.40**	.01
financial status	0.08*	.25
Moderated mediation models
Postearthquake	−0.10	−.24*	−.00
Behavioral intention		.43**
Age	−0.02**	−.04**
Age × postearthquake	−0.01 (H3)	.00 (H3)
Controls: gender	−0.42**	−.07
financial status	0.12**	.33**
Nontechnology adoption intention and behavior (i.e., control model)
Mediation model
Postearthquake	−0.17**	−.05	−.01*
Behavioral intention		.05**
Controls: gender	1.06**	−.03
financial status	0.01	.17

p < .05. **p < .01.

Discussion

Our results support H1 (i.e., the direct effect of the earthquake), whereas H2 (i.e., indirect effect of the earthquake via intention on technology adoption) and H3 (i.e., age moderation) were not supported. The results suggest that there was a decrease in technology adoption behavior after the experience of earthquake even after controlling for the effects of gender and current financial status. However, behavioral intentions did not mediate the negative effect of the earthquake on technology use, and the negative impact of the earthquake was not more pronounced in older adults. Interestingly, when technology adoption behavior and intention were replaced with nontechnology-related consumption behavior and intention, the direct effect of the disaster was not significant, suggesting that the event uniquely affected technology adoption and use.

We believe that this article is the first study that investigated the effects of experiencing a nuclear disaster on technology adoption within the context of major theoretical frameworks such as TAM and UTAUT. Those frameworks have demonstrated that researchers should take account of complex relationships among various predictors to explain technology acceptance. Previous studies found the negative impact of the disaster on specific attitude toward nuclear power (Huang et al., 2013; Siegrist & Visschers, 2013). Similarly, our results suggest that the negative impact of the disaster was applicable to adoption of advanced technology products for everyday life over and above the potential influence of gender and financial status. It is possible that the indirect experience of the nuclear meltdown in Fukushima might increase perceived risk on or lead to a loss of trust in advanced technology though this hypothesis was not testable with current survey questions. Our model also replicated a significant relationship between behavioral intention to adopt new technology and the number of technology adoptions (Davis, 1989; Venkatesh et al., 2012) although behavioral intention did not mediate the negative impact of the disaster on technology adoption.

Our model further assessed if age moderated the relationship between the earthquake and technology adoption after controlling for gender and current financial status. As shown in Table 3, consistent with previous studies (Czaja et al., 2006; Czaja & Sharit, 1998; Tacken et al., 2005), age was still negatively correlated with technology adoption. However, inconsistent with our expectation (H3), the negative impact of experience of disaster on technology adoption was not more pronounced for older adults in our sample consisting of middle-aged to older adults. One possible explanation is that our sample does not include young adults which might result in a lack of statistical power to detect age moderation across the full age range for adults.

These results were further validated in the control model where we replaced technology-related intention and behavior with nontechnology-related intention and behavior. Most importantly, the direct effect of the disaster experience on consumption of nontechnology-related products was not significant, whereas that on technology adoption behavior was significant. This result indicates that the disaster might primarily influence technology-related adoption behavior. However, the control model suggests that there was still a negative impact of the earthquake on the intention to purchase nontechnology-related products and that the intention could mediate the relationship between the earthquake and the consumption of nontechnology-related products. This negative impact on the intention to purchase nontechnology-related products might be associated with an overall change in consumerism after the Japanese had witnessed many deaths and suffering (Oishi et al., 2018).

Given that Oishi et al. (2018) also attempted to examine whether the earthquake influenced the Japanese attitudes toward new technology using the same survey, it is worth noting differences between this study and Oishi et al. (2018). This study conducted a factor analysis using all 12 attitudinal items on new products and technology in the survey, and it allowed us to generate a construct of behavioral intention first (i.e., willingness to use new technology) which has been considered as a key construct underlying theoretical mechanism that drives technology adoption behavior (Venkatesh et al., 2012). It is interesting that Oishi et al. (2017) found no significant evidence of change in consumerism and attitudes toward new technology after the disaster. However, Oishi et al.’s (2017) findings relied on participants’ responses on individual items measuring generic attitudes toward new technology or products. Those measures might not be sensitive enough to detect actual changes in attitudes toward new technology and adoption behavior. To address this limitation, this study first identified underlying constructs for the 12 attitudinal items. With the addition of a behavior variable reflecting technology adoption, this factor analysis further allowed us to take account of the relationship between behavioral intention and behavior in our models.

Limitations

There are several caveats to interpreting these findings. First, in the control model, it was possible that the fear of radiation contamination might have affected the respondents’ intention to purchase some nontechnology-related products, specifically food and groceries grown in the radiation-affected areas. This could be another possible explanation of the negative impact of earthquake on the intention to purchase nontechnology-related products. In addition, as a measure of nontechnology-related consumption, this study relied on the respondents’ self-report on change in shopping and eating out activities within the past 12 months, which might not be sensitive enough to reflect actual change in nontechnology-related consumption behavior. Such shopping activities also could include the purchase of new technology products reflected in the measure of technology adoption. These limitations of the control model could make itself less comparable to the model with technology adoption behavior. Second, as discussed in Oishi et al. (2018), the effect of experiencing the earthquake on technology adoption could also differ in populations from other countries who are not as familiar with earthquakes as the Japanese. The apparent resilience of older Japanese adults, no interaction effects for age with experiencing the earthquake, supports that interpretation. Third, it should be noted that the effect sizes of path analyses were small in terms of the amounts of variance accounted for. However, the fact that there was a measurable effect of the earthquake on technology adoption in the mediation model and that it tended to become weak in the control model is still noteworthy. Finally, this national survey did not allow us to capture a long-term effect of the earthquake (Oishi et al., 2018), and both the measures of technology adoption and nontechnology-related consumption behaviors could include purchase or consumption behaviors (i.e., shopping and eating out) that they made before the disaster occurred. These caveats suggest that it is important to see if the current findings can be replicated elsewhere using more direct measures of intentions and adoption behavior as well as a within-subject approach.

Conclusion and Implications

Although it was exploratory in nature, this study was a rare natural experiment that allowed us to directly investigate the relationships between the experience of a major disaster—which resulted in massive technology failure—and technology adoption and how they were associated with age. Our findings supported that Japanese middle-aged to older adults reported fewer technology adoption behaviors after the experience of the earthquake.

Particularly, the differing patterns of relationships for technology and nontechnology intentions and behavior suggests that future studies should pay more attention to the relationship between technology adoption and a specific loss of trust in technology systems. Although age has been negatively associated with new technology adoption, it is important to consider that older adults potentially have a great deal to gain from technology systems. A good example is the opportunity to prolong mobility through the use of advanced driving assistance systems (ADASs) and automated vehicles (AVs) when ability to drive safely wanes later in life. Older adults express higher valuation for ADAS such as blind spot detection than do younger drivers (Souders et al., 2017). Nonetheless, such artificial intelligence systems are fallible, and assuring that seniors continue to trust such systems requires that their expectations for safety gains are reasonable and that their learning cost will be minimized. It is also worth noting that people might lose trust in AV technology system after recent crashes and a rash of negative press about them (Smith, 2018).

Although the negative impact of the earthquake was not more pronounced in older adults, our results still suggest that adoption of advanced technology could be affected by the natural disaster aftermath over and above the roles of gender and financial status in technology adoption. One of the lessons learned from Fukushima is that it is not possible to envision and control for all contingencies. For seniors to trust technology, vendors need to promote realistic expectations. Researchers need to pay more attention to the issue of how changes in loss of trust and/or perceived risk affect technology adoption interacting with other relevant factors, particularly, age-related factors and abilities (Hoff & Bashir, 2015; Merkel et al., 2016).

Footnotes

Appendix

The following 12 questions refer to your attitude toward high-technology products. Please respond to each statement using the same 7-point scale:

1 = strongly disagree; 2 = disagree; 3 = disagree somewhat; 4 = neither disagree nor agree; 5 = agree somewhat; 6 = agree; and 7 = strongly agree

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Jong-Sung Yoon

References

Chen

Chan

A. H.

(2014). Predictors of gerontechnology acceptance by older Hong Kong Chinese. Technovation, 34, 126–135. https://doi.org/10.1016/j.technovation.2013.09.010

Cheron

Kohlbacher

(2018). Older consumers’ adoption of innovation in Japan: The mediating role of cognitive age. Journal of International Consumer Marketing, 30(4), 244–259. https://doi.org/10.1080/08961530.2018.1436481

Chung

J. E.

Park

Wang

Fulk

McLaughlin

(2010). Age differences in perceptions of online community participation among non-users: An extension of the technology acceptance model. Computers in Human Behavior, 26, 1674–1684. https://doi.org/10.1016/j.chb.2010.06.016

Cruz

A. M.

Steinberg

L. J.

Vetere-Arellano

A. L.

(2006). Emerging issues for Natech disaster risk management in Europe. Journal of Risk Research, 9, 483–501.

Czaja

S. J.

Charness

Fisk

A. D.

Hertzog

Nair

Rogers

Sharit

(2006). Factors predicting the use of technology: Findings from the Center on Research and Aging and Technology Enhancement (CREATE). Psychology and Aging, 21, 333–352. https://doi.org/10.1037/0882-7974.21.2.333

Czaja

S. J.

Sharit

(1998). Age differences in attitudes toward computers. The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences, 53, 329–340. https://doi.org/10.1093/geronb/53B.5.P329

Davis

F. D.

(1989). Perceived usefulness, perceived ease of use, and user acceptance of information technology. MIS Quarterly, 13, 319–340. https://doi.org/10.2307/249008

Davis

F. D.

Venkatesh

(2004). Toward preprototype user acceptance testing of new information systems: Implications for software project management. IEEE Transactions on Engineering Management, 51, 31–46. https://doi.org/10.1109/TEM.2003.822468

Dunning

(2012). Natural experiments in the social sciences: A design-based approach. Cambridge University Press.

10.

Goebel

Krekel

Tiefenbach

Ziebarth

N. R.

(2015). How natural disasters can affect environmental concerns, risk aversion, and even politics: Evidence from Fukushima and three European countries. Journal of Population Economics, 28, 1137–1180. https://doi.org/10.1007/s00148-015-0558-8

11.

Hayes

A. F.

(2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. Guilford Press.

12.

Hayes

A. F.

(2015) An index and test of linear moderated mediation, Multivariate Behavioral Research, 50, 1–22. https://doi.org/10.1080/00273171.2014.962683

13.

Hoff

K. A.

Bashir

(2015). Trust in automation: Integrating empirical evidence on factors that influence trust. Human Factors, 57, 407–434. https://doi.org/10.1177/0018720814547570

14.

Huang

Zhou

Han

Hammitt

J. K.

Liu

(2013). Effect of the Fukushima nuclear accident on the risk perception of residents near a nuclear power plant in China. Proceedings of the National Academy of Sciences of the United States, 110, 19742–19747. https://doi.org/10.1073/pnas.1313825110

15.

Kingston

(2012). Introduction. In Kingston

(Ed.), Natural disaster and nuclear crisis in Japan: Response and recovery after Japan’s 3/11 (pp. 1–11). Routledge.

16.

Kulviwat

Bruner

G. C.

II Al-Shuridah

(2009). The role of social influence on adoption of high tech innovations: The moderating effect of public/private consumption. Journal of Business Research, 62, 706–712. https://doi.org/10.1016/j.jbusres.2007.04.014

17.

Lee

Coughlin

J. F.

(2015). PERSPECTIVE: Older adults’ adoption of technology: An integrated approach to identifying determinants and barriers. Journal of Product Innovation Management, 32, 747–759. https://doi.org/10.1111/jpim.12176

18.

Lee

Moschis

G. P.

Mathur

(2001). A study of life events and changes in patronage preferences. Journal of Business Research, 54, 25–38.

19.

Mayhorn

C. B.

(2005). Cognitive aging and the processing of hazard information and disaster warnings. Natural Hazards Review, 6, 165-170.

20.

Mathur

Moschis

G. P.

Lee

(2008). A longitudinal study of the effects of life status changes on changes in consumer preferences. Journal of the Academy of Marketing Science, 36, 234–246. https://doi.org/10.1007/s11747-007-0021-9

21.

Merkel

Enste

Hilbert

Chen

Chan

A. H. S.

Kwon

(2016). Technology acceptance and aging. In Kwon

(Ed), Gerontechnology: Research, practice, and principles in the field of technology and aging (pp. 335–349). Springer.

22.

Meuter

M. L.

Bitner

M. J.

Ostrom

A. L.

Brown

S. W

. (2005). Choosing among alternative service delivery modes: An investigation of customer trial of self-service technologies. Journal of Marketing, 69, 61–83. https://doi.org/10.1509/jmkg.69.2.61.60759

23.

Ngo

E. B.

(2001). When disasters and age collide: Reviewing vulnerability of the elderly. Natural Hazards Review, 2, 80–89.

24.

Oishi

Kohlbacher

Choi

(2018). Does a major earthquake change attitudes and well-being judgments? A natural experiment. Social Psychological and Personality Science, 9, 364–371. https://doi.org/10.1177/1948550617707016

25.

Oishi

Yagi

Komiya

Kohlbacher

Kusumi

Ishii

(2017). Does a major earthquake change job preferences and human values? European Journal of Personality, 31, 258–265. https://doi.org/10.1002/per.2102

26.

Porter

C. E.

Donthu

(2006). Using the technology acceptance model to explain how attitudes determine Internet usage: The role of perceived access barriers and demographics. Journal of Business Research, 59, 999–1007. https://doi.org/10.1016/j.jbusres.2006.06.003

27.

Rehdanz

Welsch

Narita

Okubo

(2015). Well-being effects of a major natural disaster: The case of Fukushima. Journal of Economic Behavior & Organization, 116, 500–517.

28.

Rodriguez

Donner

Trainor

J. E.

(Eds.). (2018). Handbook of disaster research (2nd ed.). Springer.

29.

Sarker

Wells

J. D.

(2003). Understanding mobile handheld device use and adoption. Communications of the ACM, 46, 35–40.

30.

Schulz

Wahl

H. W.

Matthews

J. T.

De Vito Dabbs

Beach

S. R.

Czaja

S. J.

(2014). Advancing the aging and technology agenda in gerontology. The Gerontologist, 55, 724–734.

31.

Siegrist

Visschers

V. H.

(2013). Acceptance of nuclear power: The Fukushima effect. Energy Policy, 59, 112–119. https://doi.org/10.1016/j.enpol.2012.07.051

32.

Silver

R. C.

Garfin

D. R.

(2016). Coping with disasters. In Norcross

J. C.

VandenBos

G. R.

Freedheim

D. K.

Pole

(Eds.), APA handbook of clinical psychology: Psychopathology and health (Vol. 4, pp. 597–611). American Psychological Association.

33.

Smith

(2018, May 23). Distrust in self-driving cars on the rise after crashes. CNN. http://money.cnn.com/2018/05/22/technology/self-driving-cars-aaa/index.html

34.

Sobel

M. E.

(1982). Asymptotic confidence intervals for indirect effects in structural equation models. Sociological Methodology, 13, 290–312.

35.

Souders

D. J.

Best

Charness

(2017). Valuation of active blind spot detection systems by younger and older adults. Accident Analysis & Prevention, 106, 505–514. https://doi.org/10.1016/j.aap.2016.08.020

36.

Tacken

Marcellini

Mollenkopf

Ruoppila

Szeman

(2005). Use and acceptance of new technology by older people. Findings of the international MOBILATE survey: “Enhancing mobility in later life.” Gerontechnology, 3, 126–137. https://doi.org/10.4017/gt.2005.03.03.002.00

37.

Tellis

G. J.

Yin

Bell

. 2009. Global consumer innovativeness: Cross-country differences and demographic commonalities. Journal of International Marketing, 17, 1–22. https://doi.org/10.1509/jimk.17.2.1

38.

Tiefenbach

Kohlbacher

(2015a). Disasters, donations, and tax law changes: Disentangling effects on subjective well-being by exploiting a natural experiment. Journal of Economic Psychology, 50, 94–112. https://doi.org/10.1016/j.joep.2015.07.005

39.

Tiefenbach

Kohlbacher

(2015b). Happiness in Japan in times of upheaval: Empirical evidence from the national survey on lifestyle preferences. Journal of Happiness Studies, 16, 333–366. https://doi.org/10.1007/s10902-014-9512-9

40.

Venkatesh

Morris

M. G.

Davis

G. B.

Davis

F. D.

(2003). User acceptance of information technology: Toward a unified view. MIS Quarterly, 27, 425–478. https://doi.org/10.2307/30036540

41.

Venkatesh

Thong

J. Y. L.

(2012). Consumer acceptance and use of information technology: Extending the unified theory of acceptance and use of technology. MIS Quarterly, 36, 157–178.

42.

Wang

Matz-Costa

Miller

Carr

D. C.

Kohlbacher

(2018). Uses and gratifications sought from mobile phones and loneliness among Japanese midlife and older adults: A mediation analysis. Innovation in Aging, 2, Article igy027. https://doi.org/10.1093/geroni/igy027