According to the 1921 Census of Yugoslavia (i.e. before the War, before Communism, before Tito), there was a clear Muslim and Albanian majority in Kosovo. We also see this in the 1931 census and, as far as I know, all subsequent censuses except for 1991, where many Muslim boycotted (in the 2011, many Serbs boycotted--there was no attempt at a 2001 census in Kosovo, as far as I am aware).

The 1921 and 1931 censuses have people tabulated by language and religion, but not cross tabulated, so we don't know how many of the Albanians are Muslim (there is a Christian Albanian minority) and how many of the Muslims are Albanian (there are other Muslim groups, like Slavic Muslims, Romani, and Turks). Also, neither of those is self-identitified ethnicity (though I suspect in that period, the language questions was a fairly good method of determining ethnicity). However, as I mentioned, both of these show an Albanian majority, long before Tito.

There is a debate though about exactly when the region became Albanian majority is debated. The area was clearly Serbian majority in the middle ages. Some have argued it was a late as the late 19th century with the collapse of the Ottoman Empire in the Balkans, some people have argued it was as early as 16th century for parts of Kosovo. It possible (but not necessary) that both statements could be correct.

Statistical Yearbook of the Republic of Kosovo 2016, put out out by the Kosovo government obviously using statistics gathered by the Yugoslavian government 1991 and before, shows ethnicity data going back to the 1948 census on for ethnicity (in order to have an apple to apples comparison--as I mentioned above, the 1921 and 31 censuses weren't actually asking about ethnicity, but about religion and language), but you can see that already in 1948, when Tito just came to power, there was a clear Albanian majority that only gradually expanded (remember, don't trust 2011 because of the ethnic Serb boycott and note that they omitted the 1991 census because of the ethnic Albanian and Muslim boycott). Look at Tab 2.4 and Fig 2.6, both on page 28.