https://doi.org/10.5061/dryad.ttdz08m4k

The experiment was conducted at the Insect-Plant Interaction Laboratory of the Entomology Department at National Chung Hsing University in Taichung, Taiwan. In the first part, the experiment involved feeding S. manilae with different concentrations of caffeine solution and parasitizing late 1st instar S. litura larvae for five consecutive days. The experiment was conducted between 8:00 a.m. and 5:00 p.m. In the second part, the longevity of S. manilae was tested across genders with four treatments. All the experiments were conducted under controlled conditions (26 ± 2°C, 14:10 hr L:D, 60% RH). To avoid any potential influence on the next generation, all the offspring of tested parasitoids were not used for further experiment or reproduction.

Description of the data and file structure

• Both of the csv files can directly import into R program.
• Both of the files can be separated by treatment. In the treatment column, t2 represented the high dose treatment; t4 represented moderate dose and t6 represent the low dose treatment.
• The file entitled "caffeine data parasitism rate" contained detailed information about the 10 rounds experiment. The detailed information about the column: column D (number of parasitized S. litura larvae), E (number of parasitized S. litura larvae, but the cocoon was incompletely), F (number of S. litura larvae died during development stage), G (number of S. litura larvae disappeared during development stage), H (number of S. litura larvae failed to found after the parasitized process), I (the missing data), J (number of S. litura larvae not been parasitized), K (number of total left data, (30 - column F, G, H)), L (number of male offspring), M (number of female offspring), N (number of parasitoids eclosion successfully), O (parasitism rate), P (failed cocoon rate), Q (eclosion rate). The parasitism rate were calculated by (column D/ column K); cocoon rate was calculated by (1-(column E/ (column D + column E))); eclosion rate was calculated by (column N/column D); gender ratio was calculated by (column L/(column L + column M)).
• The file entitled "caffeine data 1 to 10 V2" contained the detailed information about the individual result of S. litura larvae. The letters in column E represented the status of the S. litura larvae, s (number of parasitized S. litura larvae), hs (number of parasitized S. litura larvae, but the cocoon was incompletely), g (number of S. litura larvae died during development stage), d (number of S. litura larvae disappeared during development stage), unf (number of S. litura larvae failed to found after the parasitized process), m (the missing data), x (number of S. litura larvae not been parasitized). Column F represented cocoon weight of S. manilae; column G (duration day of larval stage); column H (duration day of pupal stage); I (gender of S. manilae ). The "NA" cells found in the column G, H, I means the larva of S. litura were not growing successfully into the next stage, that is the reason why the NA cells occurred.
• Blank cells correspond to missing values.

Code/Software

Statistical analyses were conducted using R version 4.3.0 (R Core Team, 2023). The values of parasitism rate (%), pupation rate (%), eclosion rate (%), duration of the larval stage (day), pupal stage (day), and cocoon weight (mg) utilized the generalized linear model (GLM) with a quasibinomial distribution and error link logit (function “glm”) with a Tukey post-hoc test at a significance level of p<0.05 to assess differences between treatments. Sex ratio between the four treatments, Pearson's chi-squared test (function “chisq.test”) was conducted at a significance level of p<0.05. Additionally, longevity between caffeine concentrations and gender, the longevity of adult parasitoid was analyzed using Kaplan-Meier at a significance level of p<0.05 (function “survfit”).

code of glm:
 
fit1 <- glm(column header of rate~treatment, data=caffeine_parasitism,
 family = quasibinomial(link = "logit"))
summary(fit1)
library(multcomp)
fit1.t <- glht(fit1, linfct = mcp(treatment = "Tukey"))
summary(fit1.t)
 
code of pearson's chi square:
 
plot_gender<-ggbarstats(data = file,
 x=gender,
 y=treatment,
 label = "both",
 centrality.label.args=list(size=80))
plot_gender.f<-plot_gender+
 My_Theme
plot_gender.f
 
sig_level_table<-table(gender_select$treatment,gender_select$gender)
sig_level_table
pairwise_prop_test(sig_level_table)
 
code of Kaplan-Meier test:
 
male_survival_curve<- survfit(Surv(male, status) ~ concentration, data =male_survival)
 
<br>
<br>