Table 4.
Structural Equation Modeling (SEM) Results
| Dependent variable | Independent variable | Coefficients and significance† | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Variable | Type | Levels | Grocery | Other Items | Food | ||||
| Pre-pandemic in-store shopping frequency | Age | Continuous | NA | 0.11 | *** | 0.03 | 0.03 | ||
| Graduate degree | Binary | 1: Graduate degree (e.g., MS, PhD, MBA) 0: Otherwise |
−0.10 | *** | NA | −0.03 | |||
| College degree | Binary | 1: College degree (e.g., BA, BS) 0: Otherwise |
−0.06 | ** | NA | NA | |||
| Gender | Binary | 1: Female 0: Male |
NA | 0.00 | −0.20 | *** | |||
| Marital status | Binary | 1: Married 0: Otherwise |
0.04 | * | 0.04 | ** | NA | ||
| Medium income | Binary | 1: Annual household income $625K to $1.250M 0: Otherwise |
−0.05 | ** | −0.06 | *** | 0.05 | ** | |
| High income | Binary | 1: Annual household income >$1.25M 0: Otherwise |
−0.07 | *** | −0.05 | ** | 0.10 | *** | |
| Asian, African-American, or Latino race | Binary | 1: Asian, African-American, and Latino 0: Otherwise |
0.02 | −0.01 | NA | ||||
| Mixed or other races | Binary | 1: Mixed or other races 0: Otherwise |
0.03 | ** | 0.00 | NA | |||
| Pre-pandemic online shopping frequency | Age | Continuous | NA | 0.01 | −0.07 | *** | −0.12 | *** | |
| Graduate degree | Binary | 1: Graduate degree (e.g., MS, PhD, MBA) 0: Otherwise |
0.07 | ** | NA | −0.08 | *** | ||
| College degree | Binary | 1: College degree (e.g., BA, BS) 0: Otherwise |
0.09 | *** | NA | NA | |||
| Gender | Binary | 1: Female 0: Male |
NA | −0.14 | *** | −0.03 | * | ||
| Marital status | Binary | 1: Married 0: Otherwise |
−0.03 | 0.04 | NA | ||||
| Medium income | Binary | 1: Annual household income $625K to $1.250M 0: Otherwise |
−0.03 | 0.04 | * | 0.03 | |||
| High income | Binary | 1: Annual household income >$1,250K 0: Otherwise |
0.06 | *** | 0.12 | *** | 0.01 | ||
| Asian, African-American, or Latino race | Binary | 1: Asian, African-American, and Latino 0: Otherwise |
0.05 | ** | −0.06 | *** | NA | ||
| Mixed or other races | Binary | 1: Mixed or other races 0: Otherwise |
0.01 | −0.04 | ** | NA | |||
| ATS | Graduate degree | Binary | 1: Graduate degree (e.g., MS, PhD, MBA) 0: Otherwise |
−0.11 | *** | −0.15 | *** | −0.12 | *** |
| College degree | Binary | 1: College degree (e.g., BA, BS) 0: Otherwise |
−0.08 | *** | −0.12 | *** | −0.09 | *** | |
| Gender | Binary | 1: Female 0: Male |
NA | NA | −0.03 | * | |||
| Essential workers | Binary | 1: Essential worker¥ 0: Otherwise |
0.04 | * | 0.05 | ** | 0.04 | ** | |
| Asian, African-American, or Latino race | Binary | 1: Asian, African-American, and Latino 0: Otherwise |
0.14 | *** | 0.14 | *** | 0.11 | *** | |
| Mixed or other races | Binary | 1: Mixed or other races 0: Otherwise |
0.06 | *** | 0.07 | *** | 0.07 | *** | |
| Marital status | Binary | 1: Married 0: Otherwise |
−0.09 | *** | −0.07 | *** | −0.06 | *** | |
| Medium income | Binary | 1: Annual household income $625K to $1.250M 0: Otherwise |
−0.06 | *** | −0.07 | *** | −0.06 | *** | |
| High income | Binary | 1: Annual household income >$1,250K 0: Otherwise |
−0.06 | *** | −0.06 | *** | −0.06 | ** | |
| CRP | Age | Continuous | NA | 0.05 | ** | 0.05 | ** | 0.05 | ** |
| Graduate degree | Binary | 1: Graduate degree (e.g., MS, PhD, MBA) 0: Otherwise |
0.17 | *** | 0.17 | *** | 0.18 | *** | |
| College degree | Binary | 1: College Degree (e.g., BA, BS) 0: Otherwise |
0.11 | *** | 0.12 | *** | 0.12 | *** | |
| Gender | Binary | 1: Female 0: Male |
0.08 | *** | 0.08 | *** | 0.07 | *** | |
| Asian, African-American, or Latino race | Binary | 1: Asian, African-American, and Latino 0: Otherwise |
0.07 | *** | 0.07 | *** | 0.07 | *** | |
| Mixed or other races | Binary | 1: Mixed or other races 0: Otherwise |
−0.04 | ** | −0.04 | ** | −0.04 | ** | |
| Pandemic moment | Categorical | 1: Early pandemic 2: Mid pandemic 3: Late pandemic |
−0.21 | *** | −0.21 | *** | −0.21 | *** | |
| Essential workers | Binary | 1: Essential worker¥ 0: Otherwise |
NA | 0.02 | 0.01 | ||||
| Initial change in in-store frequency | CRP | Latent | NA | −0.15 | *** | −0.15 | *** | −0.20 | *** |
| Pre-pandemic in-store frequency | Continuous | NA | −0.71 | *** | −0.68 | *** | −0.76 | *** | |
| ATS | Latent | NA | 0.19 | *** | 0.15 | *** | 0.09 | *** | |
| Pandemic moment | Categorical | 1: Early pandemic 2: Mid pandemic 3: Late pandemic |
0.12 | *** | 0.19 | *** | 0.10 | *** | |
| Initial change in online frequency | CRP | Latent | NA | 0.22 | *** | 0.08 | *** | 0.03 | * |
| Pre-pandemic in-store frequency | Continuous | NA | −0.41 | *** | −0.40 | *** | −0.50 | *** | |
| ATS | Latent | NA | −0.44 | *** | −0.13 | *** | −0.04 | *** | |
| Pandemic moment | Categorical | 1: Early pandemic 2: Mid pandemic 3: Late pandemic |
−0.01 | 0.00 | −0.01 | ||||
| Expected future change in in-store frequency | Initial change in in-store frequency | Continuous | NA | −0.44 | *** | −0.42 | *** | −0.50 | *** |
| Initial change in online frequency | Continuous | NA | 0.06 | *** | 0.05 | *** | 0.06 | *** | |
| ATS | Latent | NA | 0.11 | *** | 0.06 | *** | 0.07 | *** | |
| Expected future change in online frequency | Initial change in in-store frequency | Continuous | NA | 0.02 | 0.08 | *** | 0.07 | *** | |
| Initial change in online frequency | Continuous | NA | −0.49 | *** | −0.41 | *** | −0.48 | *** | |
| ATS | Latent | NA | 0.01 | 0.14 | *** | 0.04 | *** | ||
| Goodness of fit‡ | CFI | 0.903 | 0.855 | 0.855 | |||||
| TLI | 0.951 | 0.928 | 0.928 | ||||||
| RMSEA | 0.058 | 0.062 | 0.064 | ||||||
| SRMSR | 0.073 | 0.074 | 0.081 | ||||||
Note: ATS = attitudes toward shopping; CFI = comparative fit index; CRP = COVID risk perception; RMSEA = root mean square error of approximation; SRMSR = standardized root mean square residual; TLI = Tucker–Lewis index; NA = not applicable.
†
Significance level: ‘***’p-value < 0.01, ‘**’p-value < 0.05, ‘*’p-value ≤ 0.1, ‘’p-value > 0.1.
¥
Essential workers included those working in the following industry: medical, groceries, transportation, delivery or postal services, utilities, and building maintenance or repair.
‡
Green highlight = good fit; yellow highlight = good marginal fit.