Consideration of future consequences: Preliminary evidences for a four-factor distinction

https://doi.org/10.1016/j.paid.2015.07.022Get rights and content

Highlights

  • Results of CFA with two independent samples support a four-factor model of the CFC.

  • CFC-P1 is positively related to discounting rates for large delayed rewards.

  • CFC-F1 is negatively associated with discounting rates for small and medium delayed rewards.

  • CFC-F2 is negatively associated with discounting rates for medium delayed rewards.

  • CFC-P2 and CFC-F2 moderate the magnitude effect in inter-temporal choices.

Abstract

Consideration of either future consequences or immediate consequences plays an important role in our daily decision-making. However, no general consensus has been reached as to whether the construct – consideration of future consequences – consists of one factor or multiple factors. To examine the latent structure, we conducted two studies. In Study 1, we collected data online from 494 participants with the Consideration of Future Consequences (CFC) scale, and performed confirmatory factor analyses (CFAs) of alternative models derived from previous studies. The results indicated that a four-factor model fitted the data best. In Study 2, we administrated the CFC scale and an inter-temporal choice questionnaire to another sample of 496 participants in classrooms. Cross-validation with CFA demonstrated the four-factor solution as the best fit model. In addition, the four factors were differently correlated with the discounting rate facing various rewards. Further multilevel analysis indicated that two factors among the four moderated the magnitude effect. All these findings provided evidences for a four-factor distinction in the CFC scale.

Introduction

In everyday life, people usually consider between the immediate and future outcomes of their current actions when making decisions. While some are concerned with future consequences, others focus on immediate ones. Strathman, Gleicher, Boninger, and Edwards (1994) proposed a construct of “consideration of future consequences (CFC)” to describe such a stable individual difference. They defined the CFC as “the extent to which individuals consider the potential distant outcomes of their current behaviors and the extent to which they are influenced by these potential outcomes” (Strathman et al., 1994, p. 743). They further developed and validated a 12-item CFC scale to quantify individual differences on this construct. This scale has been widely used in many areas, such as health behaviors (e.g., Adams and Nettle, 2009, Joireman et al., 2001), academic behaviors (e.g., Joireman, 1999), and financial behaviors (e.g., Joireman et al., 2010, Joireman et al., 2005).

In their original framework, Strathman et al. (1994) regarded the CFC as a unidimensional construct. At one end were people who weighted more on future consequences, and at the other end were those who attached more importance to immediate consequences. Both exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) of the CFC scale supported a single factor solution. Accordingly, most of prior studies used an average or a sum of the five future items and the seven reversed-coded immediate items as a measure of the CFC tendency (e.g., Joireman, 1999).

However, several recent studies have raised doubts about the unidimensionality of this scale. Most of them supported a two-factor structure of the CFC scale with theoretical and empirical evidences (e.g., Adams, 2012, Joireman et al., 2008, Petrocelli, 2003).

Petrocelli (2003) conducted the first systematic examination of the latent structure of the CFC scale. Based on a principal-components factor analysis with varimax rotation in a sample of 644 undergraduate students, he found two correlated factors: one, consisted of seven reverse-coded immediate items and a future item, was related to concerning immediate consequences, while the other, consisted of four future items, was related to concerning future consequences. A CFA supported the two-factor solution against the one-factor solution.

Joireman et al. (2008) suggested a slightly different two-factor model. They named one factor with five future items as CFC-Future (CFC-F), and the other factor with seven immediate items as CFC-Immediate (CFC-I). They compared the two-factor model and Strathman et al.'s (1994) one-factor model using CFA on an aggregated database of 986 respondents, and found the two-factor model had a significantly better fit to the data. This two-factor model has also been confirmed by other researchers (e.g., Adams, 2012, Toepoel, 2010). Toepoel (2010) validated the CFC with 11 waves' data (1996–2006) from a panel study in the Netherlands among a sample aged 16 and over. He used the same EFA procure as Petrocelli (2003) did, and found the two-factor solution underlying the CFC. Adams (2012) compared this two-factor model and Strathman et al.'s (1994) one-factor model in a U.K. sample of 800 participants using CFA, and found that the two-factor solution fitted the data better.

In addition, some researchers proposed other alternative models. Rappange, Brouwer, and van Exel (2009) compared Strathman et al.'s (1994) one-factor model and Petrocelli's (2003) two-factor solution with a sample of 2006 young adolescents in the Netherlands, but the results of neither model were acceptable. They then employed PCA with varimax rotation, and a multiple-factor solution emerged: one factor consisting of seven immediate items, and the other two factors consisting of the remaining future items. Hevey et al. (2010) speculated that the two-factor model might have been caused by the method effects of item wording. They compared two one-factor models (Petrocelli, 2003, Strathman et al., 1994), two two-factor models (Joireman et al., 2008, Petrocelli, 2003), and their own model (one-factor model with correlated errors) among 590 young adults, and found that their own model provided the best fit. Ryack (2012) extended the dimensionality examination of the CFC scale from college student samples to a sample of professional financial advisors, and found that a four-factor solution was supported.

So far, most studies on the latent structure of the scale have been conducted on Western samples, such as the US (e.g., Joireman et al., 2010), the U.K. (e.g., Adams, 2012), as well as the Netherlands (e.g., Rappange et al., 2009, Toepoel, 2010). Only two studies fit a specific model to data collected on Chinese samples (Liu et al., 2013, Zhang et al., 2014), but neither made systematic comparisons between all possible models. Therefore, the first aim of this study was to find out which model would be supported by adapting the CFC to the Chinese cultural context and exploring the nature of the CFC construct among two large Chinese samples.

According to Hofstede's cultural dimensions, most Western countries hold a strong short-term orientation, evidenced by focusing on the present and the past, valuing immediate need gratification, and spending; by contrast, Eastern cultures (e.g., China, Korea, and Japan) have a long-term orientation, characterized as fostering values involving future-oriented rewards, persistence, and thrift (Hofstede & Minkov, 2010). Chinese people might exhibit similar characteristics as financial advisors in Ryack's (2012) study when weighting immediate and future consequences. Therefore, we speculated that the factor structure might display a similar four-factor solution with that in Ryack's (2012) study.

Temporal choices involve a balance between long-term larger rewards and short-term smaller rewards (Joireman et al., 2008). Temporal discounting refers to the tendency of animals and people to prefer the smaller sooner rewards over the larger later rewards (e.g., Green, Myerson, Holt, Slevin, & Estle, 2004). Several studies have examined the relationship between temporal discounting and the CFC or two CFC factors. Joireman et al. (2005) found that temporal discounting was negatively related to CFC scores. In addition, Joireman et al. (2008) found that temporal discounting was positively related to CFC-I, while negatively related to CFC-F. Charlton (2011) reported temporal discounting was negatively correlated with CFC total scores, and positively correlated with CFC-I scores, but was not related to CFC-F scores.

In the abovementioned three studies, an average discounting rate (DR) was used. However, reward size affects DR, which is also referred to as magnitude effect (Loewenstein & Prelec, 1992). The larger the reward size, the lower the DR. This phenomenon is very common in many temporal discounting circumstances (Green, Fristoe, & Myerson, 1994). It is still unknown whether the CFC factors differed in the prediction of temporal discounting when people face varied sizes of rewards. Therefore, the second aim was to find whether the CFC factors played different roles in the prediction of inter-temporal choice, and examine whether they moderated the magnitude effect. Given the fact that little empirical data is available for ascertaining the mechanism underlying the magnitude effect (Loewenstein and Prelec, 1992, Loewenstein and Thaler, 1989), we were not going to formulate specific hypotheses about how the CFC factors would moderate the effect of amount on DR.

In Study 1, we compared the fits of several factor models of the scale by means of confirmatory factor analysis with a Chinese sample. In Study 2, we repeated the similar procedure with another Chinese sample, and related the CFC scores to inter-temporal choices.

Section snippets

Participants

Undergraduate and postgraduate students who attended a statistics course in a university in Nanjing, China were recruited to participate in an online study. The final sample consisted of 229 males (46.4%) and 265 females (53.6%). Their ages ranged from 16 to 35 years (M = 22. 26, SD = 2. 55). They received course credits as reimbursement.

The CFC scale

The scale was used to assess participants' concern with future consequences. It was developed by Strathman et al. (1994), and has been widely used (e.g., Joireman

Study 2

Confirmatory factor analysis of the CFC scale and linking to inter-temporal choice.

Summary

Our study explored the latent structure of the CFC with two samples in an Eastern culture, and its link to temporal choice situations. The data supported the scale to have four underlying factors, and these four factors were differently predictive of temporal discounting. Our findings provided additional evidence for the distinction between the four CFC factors, and therefore contributed to the research on individual differences in their inter-temporal choices as well as their CFC.

Factor structure of the CFC scale

The

Acknowledgments

This study was supported by the National Natural Science Foundation of China (no. 71201079), Youth Foundation of Humanities and Social Sciences, Ministry of Education of the PRC (no. 11YJC190014), Project of Philosophy and Social Sciences from the Education Department, Jiangsu Province (no. 2011SJD190004), and China Scholarship Council (no. 201306195015, no. 201506195028).

References (35)

  • S.R. Charlton

    Beyond the shadow of a trait: Understanding discounting through item-level analysis of personality scales

    The Psychological Record

    (2011)
  • L. Green et al.

    Temporal discounting and preference reversals in choice between delayed outcomes

    Psychonomic Bulletin & Review

    (1994)
  • L. Green et al.

    Discounting of delayed food rewards in pigeons and rats: Is there a magnitude effect?

    Journal of the Experimental Analysis of Behavior

    (2004)
  • G. Hofstede

    Culture's consequences: Comparing values, behaviors, institutions and organisations across nations

    (2001)
  • G. Hofstede et al.

    Long- versus short-term orientation: New perspectives

    Asia Pacific Business Review

    (2010)
  • L.T. Hu et al.

    Fit indices in covariance structure modeling: Sensitivity to underparameterized model misspecification

    Psychological Methods

    (1998)
  • L.T. Hu et al.

    Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives

    Structural Equation Modeling

    (1999)
  • Cited by (9)

    • The relationships between core values, food-specific future time perspective and sustainable food consumption

      2021, Sustainable Production and Consumption
      Citation Excerpt :

      Thus, the CFC approach seems to be highly relevant for explaining environmental and sustainable behaviour (Arnocky et al., 2014) and understanding social dilemmas and conflicts between individual and collective interests (Joireman et al., 2001, Milfont and Gouveia, 2006). Regarding the construct and operationalization of CFC, different perspectives exist regarding whether the construct consists of one, two, or multiple factors of CFCs (e.g., Zhang et al., 2015). This study accepts the two-factor structure of an individual time perspective (Arnocky et al., 2014; Joireman et al., 2008, 2012).

    • Individual differences in functional food consumption: The role of time perspective and the Big Five personality traits

      2021, Appetite
      Citation Excerpt :

      A recent meta-analysis (Murphy & Dockray, 2018) has called attention to an ongoing debate regarding the underlying factor structure of CFC. Although most studies to date have treated the CFC scale as unidimensional (Mohammed & Marhefka, 2019), increasing evidence suggests two factors (e.g., Joireman, Balliet, Sprott, Spangenberg, & Schultz, 2008; Joireman, Shaffer, Balliet, & Strathman, 2012) or even four factors (e.g., Ryack, 2012; Zhang, Kong, Zhang, & Li, 2015). One rationale for a two-factor structure is that “individuals may consider the future consequences of their actions, the immediate consequences of their actions, or both” (p. 1273).

    • On the associations between delay discounting and temporal thinking

      2019, Personality and Individual Differences
      Citation Excerpt :

      Future research should investigate the relative utility of employing general versus specific measures of temporal thinking and delay discounting to predict and understand specific behaviors of interest. Future studies could also examine whether delay discounting correlates differentially with more specific CFC dimensions (Zhang, Kong, Zhang, & Li, 2015). In conclusion, our study contributes to a growing literature examining the associations between delay discounting and temporal thinking.

    • Time perspectives and convenience food consumption among teenagers in Vietnam: The dual role of hedonic and healthy eating values

      2017, Food Research International
      Citation Excerpt :

      This study used the original version of the CFC scale (Strathman et al., 1994). Future research should use, test, adapt, and improve different scales for assessing considerations of immediate/present and future consequences, particular for young respondents (Mello et al., 2013; Zhang et al., 2012). Other personality traits can be considered in relation to consumers' time perspectives and health behaviours, such as personality traits (Carrillo, Prado-Gasko, Fiszman, & Varela, 2012).

    View all citing articles on Scopus
    1

    Contributed equally to this work and should be considered as co-first authors.

    View full text