MTRD-2DN
Luka Korenčan
2023-12-16
Podatki
Branje podatkov iz csv datoteke naložene iz Kaggle.com: https://www.kaggle.com/datasets/thedevastator/netflix-imdb-scores/
data <- read.csv("Netflix TV Shows and Movies.csv")
summary(data)## index id title type
## Min. : 0 Length:5283 Length:5283 Length:5283
## 1st Qu.:1320 Class :character Class :character Class :character
## Median :2641 Mode :character Mode :character Mode :character
## Mean :2641
## 3rd Qu.:3962
## Max. :5282
##
## description release_year age_certification runtime
## Length:5283 Min. :1953 Length:5283 Min. : 0.0
## Class :character 1st Qu.:2015 Class :character 1st Qu.: 45.0
## Mode :character Median :2018 Mode :character Median : 87.0
## Mean :2016 Mean : 79.2
## 3rd Qu.:2020 3rd Qu.:106.0
## Max. :2022 Max. :235.0
##
## imdb_id imdb_score imdb_votes
## Length:5283 Min. :1.500 Min. : 5
## Class :character 1st Qu.:5.800 1st Qu.: 521
## Mode :character Median :6.600 Median : 2279
## Mean :6.533 Mean : 23407
## 3rd Qu.:7.400 3rd Qu.: 10144
## Max. :9.600 Max. :2268288
## NA's :161. raziskovalno vprašanje: Ali obstaja statistično pomembna razlika v IMDb ocenah med TV oddajami in filmi na Netflixu?
1.1. Analiza
Pretvorba vrednosti type iz MOVIE in SHOW v 0 in 1 (vrednost MOVIE dodelimo vrednost 0, saj predstavlja modus vrednosti spremenljivke)
data$type <- factor(data$type, levels = c("MOVIE", "SHOW"), labels = c(0, 1))
head(data)## index id title type
## 1 0 tm84618 Taxi Driver 0
## 2 1 tm127384 Monty Python and the Holy Grail 0
## 3 2 tm70993 Life of Brian 0
## 4 3 tm190788 The Exorcist 0
## 5 4 ts22164 Monty Python's Flying Circus 1
## 6 5 tm14873 Dirty Harry 0
## description
## 1 A mentally unstable Vietnam War veteran works as a night-time taxi driver in New York City where the perceived decadence and sleaze feed his urge for violent action, attempting to save a preadolescent prostitute in the process.
## 2 King Arthur, accompanied by his squire, recruits his Knights of the Round Table, including Sir Bedevere the Wise, Sir Lancelot the Brave, Sir Robin the Not-Quite-So-Brave-As-Sir-Lancelot and Sir Galahad the Pure. On the way, Arthur battles the Black Knight who, despite having had all his limbs chopped off, insists he can still fight. They reach Camelot, but Arthur decides not to enter, as "it is a silly place".
## 3 Brian Cohen is an average young Jewish man, but through a series of ridiculous events, he gains a reputation as the Messiah. When he's not dodging his followers or being scolded by his shrill mother, the hapless Brian has to contend with the pompous Pontius Pilate and acronym-obsessed members of a separatist movement. Rife with Monty Python's signature absurdity, the tale finds Brian's life paralleling Biblical lore, albeit with many more laughs.
## 4 12-year-old Regan MacNeil begins to adapt an explicit new personality as strange events befall the local area of Georgetown. Her mother becomes torn between science and superstition in a desperate bid to save her daughter, and ultimately turns to her last hope: Father Damien Karras, a troubled priest who is struggling with his own faith.
## 5 A British sketch comedy series with the shows being composed of surreality, risqué or innuendo-laden humour, sight gags and observational sketches without punchlines.
## 6 When a madman dubbed 'Scorpio' terrorizes San Francisco, hard-nosed cop, Harry Callahan – famous for his take-no-prisoners approach to law enforcement – is tasked with hunting down the psychopath. Harry eventually collars Scorpio in the process of rescuing a kidnap victim, only to see him walk on technicalities. Now, the maverick detective is determined to nail the maniac himself.
## release_year age_certification runtime imdb_id imdb_score imdb_votes
## 1 1976 R 113 tt0075314 8.3 795222
## 2 1975 PG 91 tt0071853 8.2 530877
## 3 1979 R 94 tt0079470 8.0 392419
## 4 1973 R 133 tt0070047 8.1 391942
## 5 1969 TV-14 30 tt0063929 8.8 72895
## 6 1971 R 102 tt0066999 7.7 153463Po uspešni pretvorbi vrednosti bom izvedel t test
t_test_result <- t.test(imdb_score ~ type, data = data)
print(t_test_result)##
## Welch Two Sample t-test
##
## data: imdb_score by type
## t = -23.875, df = 3976.8, p-value < 2.2e-16
## alternative hypothesis: true difference in means between group 0 and group 1 is not equal to 0
## 95 percent confidence interval:
## -0.8120173 -0.6887780
## sample estimates:
## mean in group 0 mean in group 1
## 6.266980 7.017377S parametričnim preizkusom t test, sem uspešno zavrnil domnevo 0, da je povprečna ocena IMDb za TV serije in filme enaka.
Dokazal sem, da so ocene TV serij v povprečju višje ocenjene kot filmi.
Pred izvedbo neparametričnega preizkusa in primerjavo rezultatov bom predstavil še opisno statistiko s pomočjo grafa kvantilov.
boxplot(imdb_score ~ type, data = data,
main = "Graf kvantilov IMDb ocen - Film proti TV serija",
xlab = "Vrsta (0: MOVIE, 1: SHOW)",
ylab = "IMDb ocena",
col = c("red", "green"))
Iz dobljenega grafa kvantilov je očitno vidna razlika o porazdelitvi IMDb ocen med filmi in TV serijami, mediana TV serij je skoraj enaka tretjemu kvantilu IMDb ocen pri filmih.
Opravil bom še neparametričen Wilcoxonov preizkus za primerjavo dveh neodvisnih skupin.
wilcox_test_result <- wilcox.test(imdb_score ~ type, data = data)
print(wilcox_test_result)##
## Wilcoxon rank sum test with continuity correction
##
## data: imdb_score by type
## W = 1923314, p-value < 2.2e-16
## alternative hypothesis: true location shift is not equal to 0Rezultati Wilcoxonovega preizkusa podpirajo domnevo o statistično pomembni razliki med ocenami IMDb, ko gre za type MOVIE ali za type SHOW.
Ničelno domnevo lahko zavrnem pri P-vrednosti (< 0,05).
1.2. Ugotovitve
Oba preizkusa sta pripeljala do istega rezultata, da ničelno domnevo lahko zavrnem, če pa bi moral izbrati bolj primernega za dani primer, pa bi se bolj nagibal k Wilcoxonovem preizkusu, saj je slednji manj občutljiv na osamelce in ne zahteva predpostavke o normalnosti porazdelitve podatkov, prav tako je preizkus bolj robusten kar se tiče odstopanja homogenosti variance med skupinama.
2. raziskovalno vprašanje: Kako starost filma/ TV serije vpliva na IMDb ocene?
Ničelna domneva: Ni statistično pomembne povezave med starostjo filma/ TV serije in IMDb ocenami.
2.1. Analiza
Izbira/ filtriranje ustreznih stolpcev za izvedbo logistične regresije.
selected_data <- data[c("release_year", "imdb_score")]
summary(selected_data)## release_year imdb_score
## Min. :1953 Min. :1.500
## 1st Qu.:2015 1st Qu.:5.800
## Median :2018 Median :6.600
## Mean :2016 Mean :6.533
## 3rd Qu.:2020 3rd Qu.:7.400
## Max. :2022 Max. :9.600Izvedba logistične regresije s funkcijo predelane na osmih seminarjih in predavanjih - lm()
regression_model <- lm(imdb_score ~ release_year, data = selected_data)
summary(regression_model)##
## Call:
## lm(formula = imdb_score ~ release_year, data = selected_data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -5.0315 -0.6990 0.1173 0.8173 3.1010
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 39.298788 4.360640 9.012 < 2e-16 ***
## release_year -0.016254 0.002163 -7.514 6.7e-14 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 1.155 on 5281 degrees of freedom
## Multiple R-squared: 0.01058, Adjusted R-squared: 0.01039
## F-statistic: 56.46 on 1 and 5281 DF, p-value: 6.7e-142.2. Ugotovitve
Razlaga vrednosti iz izpisa povzetka regression_model: Estimate - ocena koeficientov v regresijskem modelu, Intercept predstavlja ocenjeno IMDb oceno za film iz leta 0, vrednost Estimate v vrstici release_year pa predstavlja spremembo imdb_score ob vsakem dodatnem letu. P-vrednosti sta nizki in kažeta statistično pomembnost koeficientov. Vrednost Multiple/ Adjusted R-squared kažeta na nizek odstotek imdb_score, ki ga pojasnim z izvedenim regresijskim modelom (moj model ne pojasni velikega dela variabilnosti IMDb ocen). Vrednost F-statistike ob tako nizki P-vrednosti kaže na statistično pomembnost pojasnjevalne spremenjivke.
Na podlagi pridobljenega izkaza lahko povzamem, da obstaja statistično pomembna povezava med vrednostima release_year in imdb_score. Povezava je zelo šibka in negativna. Iz dobljenih vrednosti lahko sklepamo, da se ocene IMDb s starostjo filma/ TV serije malo zmanjšajo, treba pa je poudariti, da narejen regresijski model ne pojasni velikega dela variabilnosti v ocenah.
Predstavitev regersijskega modela še z opsino statistiko s pomočjo razpršenega diagrama s regresijsko črto
library(ggplot2)
scatter_plot <- ggplot(selected_data, aes(x= release_year, y= imdb_score)) +
geom_point() +
geom_smooth(method = "lm", se=FALSE, color = "skyblue")
print(scatter_plot)## `geom_smooth()` using formula = 'y ~ x'