Page 100 - IJPS-11-5

International Journal of
Population Studies Older adults, gender, and emotions during pandemic

Saldaña (2013), two researchers conducted the analysis Table 1. Percentage of older adults reporting fear of infection
independently, reaching consensus through collaborative across three survey waves
review. Data were stored and organized using MAXQDA 1 wave 2 wave 3 wave
st
nd
rd
2020 software (VERBI, Germany).
n % n % n %
Following a data-driven inductive approach, a Not fear 190 59 190 59 181 56.2
conventional content analysis was conducted, allowing Fear 130 40.4 131 40.7 140 43.5
codes to emerge directly from participants’ verbatims to
capture the full range of cognitive and emotional responses Missing cases 2 0.6 1 0.3 1 0.3
without imposing predefined categories. In the first coding Total 322 100 322 100 322 100
cycle, preliminary codes were established; during the Abbreviations: n: Number of participants; %: Percentage of
second cycle, texts were re-read, and codes were reviewed participants.
and adjusted as needed. To facilitate broader classification,
codes were grouped into categories representing different Table 2. Intra‑case variations in fear responses across three
themes. survey waves
2.4. Statistical analysis Intra‑case variation Cases Percentage
No-No-No 114 35.4
To assess the association between categorical variables,
Pearson’s Chi-square test was employed. This statistical test Yes-Yes-Yes 70 21.7
was used to determine whether the observed differences in No-No-Yes 32 9.9
the frequencies of the variable categories were statistically Yes-No-No 26 8.1
significant. Pearson’s Chi-square value was calculated No-Yes-No 22 6.8
along with the degrees of freedom (df) and the level of No-Yes-Yes 20 6.2
significance (p-value). A statistical significance level of Yes-No-Yes 17 5.3
α = 0.05 was established. A p-value below this threshold Yes-Yes-No 17 5.3
was considered sufficient to reject the null hypothesis,
suggesting that the variables are significantly associated. Missing cases 4 1.2
In addition, the likelihood ratio test was used as a Total 322 99.9
complement to the analysis. Although both tests are highly Notes: Each row represents a pattern of responses (“Yes” indicating fear
correlated, the likelihood ratio can offer an alternative and “No” indicating no fear) across the waves, with the corresponding
approach to interpreting the relationship between the number of cases and their percentage distribution. The total percentage
is slightly off 100 because of rounding. Missing cases include responses
variables. The analysis included a total of 322 valid cases. that were not provided or were invalid.
The statistical analysis was performed with IBM SPSS
Statistics version 29, which automatically computed the participants reported fear at some point. Persistent fear
relevant statistics for hypothesis testing. (Yes-Yes-Yes) was observed in 22% of cases, while 41%
3. Results experienced fear intermittently. These fluctuations in fear
align with the natural variability of emotions over time.
3.1. Descriptive analysis from quantitative and To further explore the associations between fear and
qualitative data demographic and situational variables, we conducted Chi-
Table 1 illustrates the proportion of older adults who reported square tests of independence. Table 3 presents the results,
fear of infection across three survey waves. A stable 40% of highlighting that gender was the only variable showing a
participants expressed fear in the first two waves, with a slight statistically significant association with fear (p < 0.05). This
increase to 43.5% in the third wave. This rise may reflect finding underlines the role of gender in shaping emotional
shifts in the perceived threat level or pandemic context by responses during the pandemic, as previously discussed.
the third wave. Notably, these fear levels are lower than those Other variables, including age group, self-perceived
found in other Argentine studies among adults aged 18 and health, and living arrangements, did not exhibit significant
older, which could suggest distinct perceptions of risk or associations with fear.
coping mechanisms among older adults. Throughout the Building on the findings from Table 3, which identify
study, response rate remained consistent (n = 322), attesting gender as the only variable significantly associated with the
to the validity of cross-wave comparisons. perception of fear, we directed our attention to examining
Table 2 provides insight into the intra-case variation intra-case variability across the three waves, with the
in fear responses across three waves, showing that 63% of information disaggregated in terms of gender. Figure 1

Volume 11 Issue 5 (2025) 94 https://doi.org/10.36922/ijps.5686